biogeography and the evolution of coral reef fish species and the evolution of coral reef fish...
TRANSCRIPT
This file is part of the following reference:
Hodge, Jennifer Reneé (2014) Biogeography and the evolution of coral reef fish species. PhD thesis, James
Cook University.
Access to this file is available from:
http://researchonline.jcu.edu.au/42130/
The author has certified to JCU that they have made a reasonable effort to gain permission and acknowledge the owner of any third party copyright material
included in this document. If you believe that this is not the case, please contact [email protected] and quote
http://researchonline.jcu.edu.au/42130/
ResearchOnline@JCU
Biogeography and the evolution of coral reef fish
species
Thesis submitted by
Jennifer Reneé Hodge (MappSc), James Cook University, Australia
April 2014
for the degree of Doctor of Philosophy in Marine Biology
School of Marine and Tropical Biology, and
ARC Centre of Excellence for Coral Reef Studies,
James Cook University, Townsville, Queensland, Australia
i
Statement of Access
I, the undersigned, author of this thesis, understand that James Cook University will make this
thesis available for the use within the University Library and, via the Australian Digital Thesis
Network, for use elsewhere.
I understand that as unpublished work a thesis has significant protection under Copyright Act
and;
I do not wish to put any further restriction on access to this thesis.
__________________________________ ___________________
Signature Date
Jennifer Hodge
ii
Statement of Sources
I declare that this thesis is my own work and has not been submitted in any form for another
degree or diploma at my university or other institution of tertiary education. Information
derived from the published or unpublished work of others has been acknowledged in the text
and a list of references is given.
__________________________________ ___________________
Signature Date
Jennifer Hodge
iii
Electronic Copy Declaration
I, the undersigned and author of this work, declare that the electronic copy of this thesis
provided to James Cook University Library is an accurate copy of the print thesis submitted,
within the limits of technology available.
__________________________________ ___________________
Signature Date
Jennifer Hodge
iv
Statement on the Contribution of Others
This thesis includes some collaborative work with my supervisors, Professor David
Bellwood and Dr. Lynne van Herwerden, and their former PhD candidate Charmaine Read.
While undertaking these collaborations, I was responsible for the project concept and design,
data collection, analysis and interpretation, and the final synthesis of results into a form suitable
for publication. My supervisors provided intellectual guidance, equipment, access to their
sample collections, financial support, and editorial assistance. My collaborator, Charmaine
Read, performed laboratory analysis and provided the resultant molecular sequence data I
analysed in the first two data chapters of this thesis. These chapters form collaborative
publications and all authors contributed to earlier drafts of the papers.
Financial support for the project was provided by the Australian Research Council Centre
of Excellence for Coral Reef Studies. Tuition and stipend support were provided by an
Australian Postgraduate Award and the James Cook University Graduate Research Scheme.
Financial support for conference travel was provided by the Australian Research Council Centre
of Excellence for Coral Reef Studies, the Australian National Network in Marine Science
(ANNiMS), James Cook University Graduate Research Scheme, and the Australian Biological
Resources Study (ABRS) Student Travel Bursary Program.
v
Acknowledgements
Firstly, I would like to thank my supervisors, David Bellwood and Lynne van Herwerden.
Lynne, thank you for having the courage to accept me as your student despite my minimal
background in molecular biology. Thank you for guiding me through the forest of seemingly
endless phylogenetic trees early on and for believing that I was capable of finding my way out
with a thesis in hand. Your encouragement and support throughout has kept me motivated, and
your thoughtful discussions and suggestions have enhanced my work. David, thank you for
your honesty, high expectations and inspiring discussions. Your supervision has continually
improved my work and my future potential as a research scientist. Thank you both for
mentoring me over the past four years. You have each played an integral role in my
professional development, and for that I am truly grateful.
I would especially like to thank Peter Cowman for his supportive guidance. Pete, your
patient advice and consultation have helped me immensely. There are many people who have
provided support and helpful direction throughout my candidature. Thank you to Jenn Tanner,
Chris Goatley, Howard Choat, Blanche Danastas, Danielle Dixson, Andrew Hoey, John Horne,
Charlotte Johansson, Jess Maddams, Melanie Trapon and Justin Welsh. I appreciate your
friendly conversation and assistance. A special thank you is extended to Simon Brandl for
prying me away from my computer and insisting I introduce him to a tiger shark. Thank you to
all of those who lived in the TBOD for accepting me into your lab and letting me share your
space.
Thank you Colin Wen for ‘fixing’ my Dell laptop in the first few months of my
candidature. Without your unintentional removal of an essential partition I would not have been
forced to find a new laptop. Following that, I would like to thank my MacBook Pro for
introducing me to the world of seamless computing and for making this thesis and the bulk of
vi
my research possible. Thank you to those people who support and maintain the CIPRES super
computing facility and to those who have developed the open source databases and computer
programs that have afforded me the ability to access and analyse data, and find information on
how to do so freely. Thank you to G. Bernardi and K. Ma for sharing their data that had not yet
become publicly available. I am also grateful to the organisers and participants of the Adelaide
Phylogenetics Workshop 2010 and the Bodega Bay Phylogenetics Workshop 2011, hosted by
UC Davis, for playing a vital role in developing my theoretical knowledge of the field and my
ability to appropriately address the questions I sought to incorporate in my research. In
particular, I would like to thank P. Wainwright, T. Heath, T. Near and J. O’Donnell for helpful
discussions during the Bodega workshop and for their continued correspondence afterwards.
This research and thesis were made possible through generous financial support. I thank the
Commonwealth government of Australia for providing me with an Australian Postgraduate
Award. I am appreciative of funds received from the James Cook University Internal Research
Awards, the Graduate Research Scheme, and the Australian Research Council Centre of
Excellence for Coral Reef Studies. I am grateful for the receipt of additional funds for
conference travel provided by the Australian Research Council Centre of Excellence for Coral
Reef Studies, the Australian National Network in Marine Science (ANNiMS), the Australian
Biological Resources Study (ABRS) Student Travel Bursary Program, and the James Cook
University Graduate Research Scheme. A special thank you to O. Bellwood, D. Bellwood, Y.
Everingham, J. Sheaves and L. van Herwerden for employing me as a tutor during my
candidature. The training and experience I gained as a result are invaluable.
I thank the dedicated staff of the School of Marine and Tropical Biology and the Graduate
Research School for logistical support and the helpful answers they have provided. Special
thanks to J. Wilson, D. Ford, S. Francis, K. Wood, and J. Bull for their assistance. Thank you to
the staff of Orpheus Island and Lizard Island Research Stations for facilitating sample
collection, and to F. Welsh and N. Konow for the provision of additional tissue samples. A very
vii
special thank you to C. Read for all of the hard work you performed in the lab and for allowing
me to work with the resultant data.
I would like to extend a warm thank you to my friends and family in Australia and abroad
for their unwavering support and encouragement. Thank you to my parents for instilling in me
the belief that I can achieve anything I set my mind to. Thank you for supporting me even when
I set my mind to things in far away places, and when I come to visit indefinitely. Thank you to
my parents-in-law for your love and support, and for the gift of our espresso machine – I am not
sure this thesis would have come into being without it. Thank you to Andrea and Steve, Lucy
and Stef, Matt, Lockie, Melanie, Brittany and Jade for your true friendship and shared drinks.
Thank you to Monica and Dave for letting me science with you; you showed me how it should
be done. Thank you to the RBT for giving me a hobby and a festive break from research once a
month.
To my husband Stephen, your patience, kindness and encouragement have seen me through
all of the highs and lows of this journey. Thank you for understanding me and for always being
able to make me laugh. Thank you for supporting me and for believing in me even when I did
not believe in myself. Thank you for loving me. Now let’s go on another adventure!
viii
Abstract
This thesis examines the evolution of coral reef fish species, specifically the chronology
and geography of extant species divergence, and the evolutionary processes that have shaped
contemporary patterns of biodiversity. The evolutionary histories of species belonging to
multiple genera from four common coral reef fish families: the Chaetodontidae
(butterflyfishes), Labridae (wrasses), Pomacanthidae (angelfishes), and Epinephelidae
(groupers) were reconstructed based on molecular data. Resultant phylogenies were temporally
calibrated using palaeontological data. The reconstructed chronograms were combined with
detailed distributional data to determine how closely related species are geographically
distributed, and to explore the processes responsible for contemporary patterns of reef fish
diversity. First, the diversification of endemic species was explored by considering a case study
of the wrasse genus Anampses. A second case study assessed the evolution of sympatric species
within the angelfish genus Pomacanthus. Finally, a multi-family phylogenetic hypothesis was
constructed to broaden the generality of conclusions drawn from the case studies. This
expanded phylogenetic hypothesis was used to critically evaluate traditional methods of
phylogenetic age estimation; to compare the ages of species from different biogeographical
areas; and to evaluate the role of geography in the speciation of coral reef fishes. Together,
these studies have identified common evolutionary and biogeographical patterns among reef
fish species, and begun to unravel potential processes involved in species divergence and
maintenance.
A chronogram of the genus Anampses identified diversification of extant species from the
mid-Miocene onward. Evolutionarily, this resulted in a high proportion of endemic species with
varied divergence times and distributions largely restricted to the range edges of Indo-Pacific
coral reefs. Evolutionary relationships within the genus, combined with limited spatial and
ix
temporal concordance among endemics, suggest that successive peripheral speciation, or
peripheral budding, may have generated substantial species diversity within this genus. The
findings highlight the importance of peripherally isolated locations in creating and maintaining
endemic species.
Extant species in the genus Pomacanthus showed similar timing in their divergence, from
the mid-Miocene onward. In contrast to Anampses, this genus consists of species that are
largely sympatric, where 80% of sister-species demonstrated complete or substantial (> 85%)
distributional overlap. Splits between lineages within the phylogeny corroborated key
biogeographical events including the Terminal Tethyan Event and the rise of the Isthmus of
Panama, suggesting that allopatric speciation impacted the early evolutionary history of the
genus. Age-range correlation analyses revealed no significant relationship between the degree
of distributional overlap and divergence time, demonstrating that exceptional sympatry is not
restricted to evolutionarily older species. This work emphasizes the need to disentangle process
from pattern by demonstrating that a number of speciation modes, including sympatry and
peripatry, likely led to the divergence of species with exceptionally high distributional overlap.
Commonly, divergence time estimates from the nodes of a temporally-calibrated phylogeny
are used as indicators of extant species’ ages. However, this method can sometimes produce
misleading age estimates, specifically in the presence of extinction and ancestral persistence. A
method to minimize the impacts of extinction and ancestral persistence on divergence time
estimation was established. The method focuses on recent divergences (using a sister-species
approach) and involves the combination of minimum divergence time estimates (as indicators
of species’ ages) with the minimum geographical range area between two sister-species, for all
sister-species pairs. When applied to coral reef fishes, this method revealed a general pattern of
geographical range expansion with increasing evolutionary age. The differences in the trends
recovered from excluding potential biases associated with ancestral persistence (i.e. maintaining
a large geographic range over time) suggest that ancestral persistence may be prevalent among
coral reef fishes, with successive peripheral speciation impacting age-area relationships. The
x
described method may reveal the occurrence of successive peripheral speciation events across a
broad range of taxa.
The multi-family phylogeny revealed similar temporal patterns of coral reef fish divergence
among major marine realms and regions, despite differing geological histories. The
evolutionary age of most coral reef fish species ranges from 1 to 5 Ma. Notable differences
were recorded in the timing of divergence and spatial relationships of endemic species in the
Red Sea and Hawaiian Islands. Red Sea endemics have diverged consistently throughout the
past 16 Myr, whereas endemic species colonized the Hawaiian Islands in two distinct waves (0–
3 Ma and 8–12 Ma). These results suggest that markedly different processes have shaped
patterns of diversification in two prominent, peripherally isolated locations.
Important areas of common overlap and vicariance were identified through the continued
application of a sister-species approach using the multi-family phylogeny. Congruent vicariance
was detected across six previously described biogeographical barriers: the Amazon and Orinoco
barriers, Isthmus of Panama, Hawaiian Archipelago, Indo-Pacific, and a previously unnamed
barrier I term the Mid-Indian Ocean Barrier (MIOB). The MIOB is hypothesized to be driven
by the unusually high sediment content of the Ganges and Indus river systems and the resultant
impacts on physical oceanography. A high concentration of distributional overlap was strikingly
concordant with the Coral Triangle. This may suggest that the Coral Triangle harbours
sufficiently complex environments to facilitate reproductive isolation through niche
specialisation that permits closely related species to co-occur. However, the significantly lower
than expected degree of distributional overlap among sister-species in this region indicates that
it is, more frequently, an area of secondary contact between species that largely occupy adjacent
ocean basins, supporting the Centre of Overlap hypothesis. This information helps to illuminate
the roles various biogeographical regions and boundaries play in creating and maintaining
extant species diversity.
xi
In summary, coral reef fish species have an intricate evolutionary history involving a
combination of evolutionary processes that have led to the establishment of complex
contemporary biogeographical patterns. Fluctuations in soft biogeographical barriers, founder
events and potential divergence in sympatry appear to have driven present day biodiversity. The
Coral Triangle not only harbours a unique richness of species but it also supports the co-
existence of numerous sister-species. As a whole, this thesis provides a detailed description of
the temporal evolution of coral reef fish species, their contemporary geographical distributions,
and the evolutionary processes that have likely shaped their distinctive patterns of biodiversity.
xii
Table of Contents
Page
Statement of Access i
Statement of Sources ii
Electronic Copy Declaration iii
Statement on the Contribution of Others iv
Acknowledgements v
Abstract viii
Table of Contents xii
List of Figures xv
List of Tables xviii
Chapter 1: General Introduction 1
Chapter 2: General Methods 7
2.1 Case Studies 7
2.2 Multi-Family Phylogeny 12
Chapter 3: The role of peripheral endemism in species diversification: evidence from the genus Anampses (Family: Labridae) 15
Published in Molecular Phylogenetics and Evolution 62: 653–663
3.1 Introduction 15
3.2 Materials and Methods 18
3.3 Results 19
3.4 Discussion 22
Chapter 4: Evolution of sympatric species: a case study of the coral reef fish genus Pomacanthus (Pomacanthidae) 36
Published in Journal of Biogeography 40: 1676–1687
4.1 Introduction 36
4.2 Materials and Methods 39
4.3 Results 40
4.4 Discussion 45
xiii
Page
4.5 Conclusions 55
Chapter 5: On the relationship between species age and geographical range in reef fishes: are widespread species older than they seem? 56
In press Global Ecology and Biogeography
5.1 Introduction 56
5.2 Phylogenetic Methods of Age Estimation 58
5.3 Quantifying Species Age in Reef Fishes 59
5.4 Applications and Implications 74
Chapter 6: Temporal evolution of coral reef fishes: global patterns and disparity in isolated locations 77
Published in Journal of Biogeography 41: 2115–2127
6.1 Introduction 77
6.2 Materials and Methods 79
6.3 Results 82
6.4 Discussion 88
6.5 Conclusions 95
Chapter 7: The geography of speciation in coral reef fishes 97
Submitted for publication
7.1 Introduction 97
7.2 Materials and Methods 101
7.3 Results 103
7.4 Discussion 105
7.5 Conclusions 119
Chapter 8: Concluding Discussion 120
References 125
Appendix A: Supplementary materials for Chapter 2 156
Appendix B: Supplementary materials for Chapter 3 178
Appendix C: Supplementary materials for Chapter 4 184
xiv
Page
Appendix D: Supplementary materials for Chapter 5 190
Appendix E: Supplementary materials for Chapter 6 200
Appendix F: Supplementary materials for Chapter 7 203
Appendix G: Publications arising from thesis 206
xv
List of Figures
Page
Figure 3.1 A phylogeny of Anampses represented by the best maximum likelihood tree
obtained from 10 independent runs in GARLI. 21
Figure 3.2 A chronogram illustrating the temporal diversification of Anampses species. 25
Figure 3.3 The geographical distributions of Anampses species. 26
Figure 3.4 A schematic diagram of the: (a) ‘successive division’, (b) ‘successive
colonization’, and (c) ‘peripheral budding’ models of diversification observed
among Anampses species. 31
Figure 4.1 A chronogram illustrating the temproal diversification of Pomacanthus
species. 43
Figure 4.2 The geographical distributions of Pomacanthus species in the west Tethyan
clade. 44
Figure 4.3 The geographical distributions of Pomacanthus species in the east Tethyan
clade. 47
Figure 4.4 Areas of species’ range overlap within the Pomacanthus genus. 48
Figure 4.5 Plots of the relationship between degree of sympatry and node age based on
the Pomacanthus phylogeny. Two methods were used to estimate the
geographical ranges of higher clades, (a) the union method, and (b) independent
comparisons. 51
Figure 4.6 Plot of the relationship between degree of symmetry and node age based on
the Pomacanthus phylogeny. 52
Figure 4.7 Proportion of sympatric sister-species among marine taxa for which there
exists a species-level phylogeny with near complete taxon sampling (> 85%)
and more than two sister-species pairs. 54
xvi
Page
Figure 5.1 Schematic diagram representing methods of (a) temporally-calibrating nodes
from a phylogeny using molecular data, and (b) how the estimates are used in
examining the relationship between species age and geographical range. 60
Figure 5.2 Relationship between species age and geographical range plotted as
described in Figure 5.1 from an analysis of 293 reef fish species from four
families: Chaetodontidae, Labridae, Pomacanthidae, and Epinephelidae. 61
Figure 5.3 Schematic diagram illustrating the effects of extinction, or missing extant
taxa, on phylogenetic estimates of divergence time. 62
Figure 5.4 Plot of the relationship between minimum divergence times mean
geographical range area, calculated for all sister-species pairs. 65
Figure 5.5 Schematic diagram illustrating (a) the peripheral budding model of
diversification, and (b) the topological relationships resolved from a molecular
phylogeny of species that have diverged under the peripheral budding model. 67
Figure 5.6 Plot of the relationship between minimum divergence time and minimum
geographical range area using a sister-species approach. 69
Figure 5.7 Plot of the results of phylogenetically independent contrasts performed to
evaluate the consistency of the relationship between species age and (a) average
geographical range areas, and (b) randomised geographical ranges for all sister-
pairs. 71
Figure 5.8 Plot of the results of phylogenetically independent contrasts performed to
evaluate the consistency of the relationship between species age and minimum
geographical range for all sister-pairs. 72
Figure 5.9 Incorporates geographical range symmetry into the model relating minimum
geographical range to species age. 74
Figure 5.10 The relationship between species age and the degree of geographical range
symmetry. 75
xvii
Page
Figure 6.1 Biogeographical variation in the ages of coral reef fish species. (a) Map
showing two marine realms, the Indo-Pacific and Atlantic, and the regions
within each realm differentiated by colour. Age comparisons of coral reef fish
species among marine realms and regions using (b) a full-phylogeny approach,
and (c) a sister-species approach. 85
Figure 6.2 Frequency distribution of age estimates for coral reef fish species across the
entire mulit-family phylogeny. 86
Figure 6.3 Frequency distributions of age estimates for coral reef fish species (a)
endemic to the Red Sea, and (b) endemic to the Hawaiian Islands. 87
Figure 6.4 Estimated ages of Red Sea and Hawaiian endemic coral reef fish species
coded by their topological status and distributional relationships. 89
Figure 7.1 Schematic showing how (a) overlap and (b) vicariance were mapped for
sister-species pairs. 103
Figure 7.2 Patterns of vicariance among sister-species of coral reef fishes. (a)
Previously described biogeographical barriers. (b) Concordant splits between
allopatric sister-species in relation to biogeographical barriers. (c) Plot of the
average degree of distributional symmetry for allopatric sister-species. 107
Figure 7.3 Patterns of sympatry among sister-species of coral reef fishes. (a) Heatmap
showing the concentration of overlap among sympatric sister-species. (b)
Enlarged view of the Central Indo-Pacific region showing the highest
concentration of overlapping areas. (c) Plots of the average degree of sympatry
and distributional symmetry for sympatric sister-species. 109
xviii
List of Tables
Page
Table 3.1 Gene-specific models of evolution of Anampses fishes selected under the
corrected Akaike information criterion in JMODELTEST. 19
Table 4.1 Gene-specific models of evolution of Pomacanthus fishes selected under the
corrected Akaike information criterion in JMODELTEST. 41
Table 6.1 Sample sizes and resulting mean species age for two approaches used to
compare the ages of coral reef fish species between marine biogeographical
realms and regions, the (a) full-phylogeny approach and the (b) sister-species
approach. 84
1
Chapter 1: General Introduction
Understanding the complex history of earth’s endless forms, how they came to be and how
they are sustained, has been a fundamental facet of biology since the dawn of evolutionary
theory (Darwin, 1859; Wallace, 1889). Over the past several decades, the use of molecular data
has advanced the exploration of evolution and biogeography (Mantooth & Riddle, 2011;
Posadas et al., 2013; Morlon, 2014). Molecular based phylogenetic hypotheses provide a
window into the evolutionary history of living forms, detailing organismal relationships and the
temporal aspect, or chronology, of divergence. In the process of unravelling complex
biogeographical histories the chronology of divergence is a vital component because it provides
an opportunity to identify palaeogeographical events and their likely influence on the
divergence of particular taxa. Together, the chronology and geography of evolution form a
spatiotemporal framework that facilitates further investigation into the mechanistic processes of
evolution.
Because evolution is a dynamic, continuous process, phylogenetic methods can be applied
to a range of taxonomic levels. Biogeography can also be considered on an array of spatial
scales. However, the diversity of life is most often described using species as the focal unit. It is
therefore critically important to establish an evolutionary biogeographical framework at the
species level to fully understand key factors that shape diversity and the processes through
which diversity is generated. Successful projection of the biogeographical consequences of
global environmental change critically hinge on this foundational knowledge (Pimm & Raven,
2000; Hughes et al., 2003).
Geographical areas of concentrated diversity, or biodiversity hotspots, provide ideal
opportunities to study evolution and biogeography. In the marine realm, the most prominent
biodiversity hotspot, often referred to as the Coral Triangle, is centred in the Malay Archipelago
2
between the Indian and Pacific Oceans (Rosen, 1984; Hoeksema, 2007; Bellwood & Meyer,
2009). The area is a hotspot of diversity for many marine organisms including plants,
invertebrates, and coastal reef-associated fishes (reviewed in Hoeksema, 2007; Renema et al.,
2008; Bellwood et al., 2012). Of the organisms aligned with this hotspot, the reef fishes are
among the most diverse and have well-described, interesting distributional patterns, providing
an exemplary system to study the biogeography and evolution of tropical species.
Early evolutionary studies of reef fishes used traditional morphological systematics to
establish important biogeographical patterns among closely related taxa (Vari, 1978; Woodland,
1983; Winterbottom et al., 1984; Springer, 1988; Blum, 1989; Hastings, 1990). Methodological
advances in the collection and analysis of molecular sequence data led to the implementation of
molecular phylogenies as a tool to investigate the evolutionary history of reef fishes. Initial
molecular based studies were limited to a small number of taxa in select genera (Lacson &
Nelson, 1993; McMillan & Palumbi, 1995; Streelman et al., 2002; Crow et al., 2004).
However, they confirmed that the results of molecular phylogenetic approaches often aligned
with predictions made by morphological studies. Even when the two approaches differed, these
early studies reinforced their complementarities and demonstrated the effectiveness of their
combined usage to investigate biogeographical patterns.
Further phylogenetic studies clarified the taxonomic structure and distinction of reef fishes,
providing the foundational knowledge necessary to approach questions of biogeography
(Streelman & Karl, 1997; Kuriiwa et al., 2007; Chiba et al., 2009; Cooper et al., 2009; Near et
al., 2013; Sorenson et al., 2013). The development of analytical methods that combined
molecular data with palaeontological records created the opportunity to use fossils and
palaeogeographical events as calibration tools to establish a temporal framework for
phylogenetic hypotheses. While time-trees only provide estimates of the timing of divergence
events surrounded by a certain level of uncertainty, they have successfully described the tempo
and mode of divergence among reef fishes and corroborated the timing of lineage divisions with
key palaeogeographical events such as the closure of the Tethys Seaway and the rise of the
3
Isthmus of Panama (Bellwood et al., 2004; Barber & Bellwood, 2005; Robertson et al., 2006;
Fessler & Westneat, 2007; Teske et al., 2007; Cowman & Bellwood, 2011; Lin & Hastings,
2013). Temporal estimates for dominant reef fish families showed that most generic lineages
were in place by the early Miocene; however, these studies provided few direct estimates of the
timing of species divergence.
Recent comparative phylogenetic studies have modelled ancestral lineages to infer patterns
of origination and dispersal (Cowman & Bellwood, 2013a). They revealed that the Indo-
Australian Archipelago has played a number of roles in accumulating, maintaining and
exporting reef fish lineages through time. Reconstructing ancestral vicariance showed that both
hard and soft barriers could have strong effects on the evolution of reef fish lineages (Cowman
& Bellwood, 2013b). These findings illuminate the processes important in lineage
diversification among reef fishes and provide a vital frame of reference for further investigation
into their evolutionary history. However, to clarify the processes responsible for patterns of
contemporary species diversity it is necessary to develop a framework that describes the
evolution and biogeography of reef fish species.
Equally important findings have come from studies of population-level relationships within
species. These studies have highlighted key biogeographical barriers to genetic connectivity of
populations (Muss et al., 2001; Rocha, 2003; Bernardi & Lape, 2005; Gaither et al., 2010),
areas of population overlap (Bowen et al., 2001; Gaither et al., 2011a), and potentially
important modes of speciation for reef fishes (Rocha et al., 2005; Craig et al., 2006; Winters et
al., 2010). Comparative phylogeographic studies have constructively generalised intra-specific
biogeographical patterns (Hellberg, 2009; Drew & Barber, 2012; Altman et al., 2013; DiBattista
et al., 2013). If similar processes are responsible for driving genetic diversification within and
between species, comparative phylogeographic studies may serve as a tool for predicting
evolutionary and biogeographical patterns among species.
4
Phylogenetic studies have detailed the evolution of species for a growing number of reef
fish genera (McCafferty et al., 2002; Bernardi et al., 2004; Klanten et al., 2004; Read et al.,
2006; Bernardi et al., 2008; Crow et al., 2010; Choat et al., 2012; Gaither et al., 2014).
However, few studies have considered the evolutionary history of reef fish species within a
temporal and spatial framework to explore their biogeographical patterns (but see Bernardi et
al., 2008). Furthermore, comparative phylogenetic studies of reef fish species have been limited
by methodological inconsistencies of estimated divergence times among closely related species,
or the lack of temporal estimates for some taxa. Without a temporal framework describing the
chronology of divergence, it is difficult to evaluate the importance of palaeogeographical events
in generating and maintaining species. However, these comparative studies have established
broad biogeographical patterns and highlighted potentially important processes that enable
specific questions pertaining to the evolution and biogeography of coral reef fish species to be
addressed.
With the increasing availability of molecular sequence data and the acquisition of
previously un-sampled taxa it is now possible to explore the evolutionary and biogeographical
history of species within a number of reef fish genera. Within the context of the evolutionary
biogeography of ancient reef fish lineages and present-day populations, the research herein
seeks to identify common phylogenetic and distributional patterns among species to uncover
those that are likely to have shared a common biogeographical history, and to explore the
potential evolutionary and biogeographical processes involved in species divergence and
maintenance.
Aims
The primary objective of this thesis was to establish the chronology of extant species
divergence and combine it with detailed contemporary biogeographical data to form the
spatiotemporal framework necessary to investigate evolutionary processes. This was
5
accomplished through the reconstruction of the evolutionary histories of reef fish species using
molecular phylogenetic methods. Molecular sequence data for species within multiple coral reef
fish genera were generated or mined from existing data. A range of recent phylogenetic
reconstruction and age estimation programs were used to analyse the molecular data in
combination with palaeontological data. The resultant phylogenetic hypotheses, along with
further biogeographical approaches, were used to address specific objectives relating to the
following key questions:
1. When did reef fish species diverge?
2. How are closely related species geographically distributed? and
3. What does this suggest about the processes responsible for contemporary patterns of
reef fish diversity?
These key questions are addressed in five distinct chapters outlined below. Chapters 3, 4, and 6
correspond directly to publications arising from this thesis (see Appendix G).
Thesis outline
Question 1 was addressed in all five chapters by employing the latest Bayesian inference
methods to reconstruct temporally-calibrated phylogenetic hypotheses based on molecular and
palaeontological data for genera with complete (or near complete) sampling of extant, nominal
species. Question 2 involved the geographical analysis of contemporary distributions of species
in conjunction with their estimated phylogenetic relationships and divergence times. Together,
this information provided the framework necessary to explore variations in Question 3.
Chapter 3 investigated the role of peripheral isolation in the diversification of reef fish
species in a case study of the wrasse genus Anampses (Family: Labridae). Chapter 4 examined
the evolution of sympatric species within the angelfish genus Pomacanthus (Family:
6
Pomacanthidae). To explore the evolutionary history of reef fish species more broadly,
Chapters 5–7 examined inter-specific relationships of 53 genera from four major coral reef fish
families (Chaetodontidae, Labridae, Pomacanthidae, and Epinephelidae). Chapter 5 critically
evaluated methods commonly used to analyse the relationship between species age and
geographical range area by addressing the inherent effects of phylogenetic age estimation using
coral reef fishes as an example. Chapter 6 explored global patterns in the temporal evolution of
reef fish species, with particular focus on patterns of divergence among species restricted to
isolated locations. Chapter 7 analysed the geography of speciation in reef fishes where
common areas of sympatry and vicariance among sister-species were identified.
Finally, Chapter 8 summarised the preceding chapters, providing an overview of the
evolutionary history of coral reef fish species and the processes potentially important in
generating such exceptional species diversity.
7
Chapter 2: General Methods
2.1 Case Studies
Data selection
Data for both case studies (Chapters 3, 4) were collected in the form of tissue samples taken
from all extant, nominal species corresponding to the applicable genera, Anampses and
Pomacanthus (12 and 13 species, respectively; nomenclature following Randall, 1972, 1986;
Allen et al., 1998; Bellwood et al., 2004). Tissues of Anampses viridis were unavailable. This
species, endemic to Mauritius, has not been observed in recent years, despite dedicated
sampling efforts. It is thought to be extinct (Hawkins et al., 2000; Dulvy et al., 2003), or to
have been taxonomically confused (Russell & Craig, 2013). The closest sister groups to
Anampses and Pomacanthus are not known with certainty; however, recent phylogenetic studies
of the family Labridae (Westneat & Alfaro, 2005; Kazancıoğlu et al., 2009; Cowman &
Bellwood, 2011) and Pomacanthidae (Bellwood et al., 2004) were consulted to identify putative
sister taxa. These studies identified Macropharyngodon and Halichoeres as sister to Anampses,
and Chaetodontoplus, Pygoplites, and Holacanthus as sister to Pomacanthus. Representative
species from each sister genus were included in the phylogenetic analyses, as well as more
distant species used to root the phylogenies and provide nodes for fossil calibration (see Tables
S2.1 and S2.2 in Appendix A, and subsection Age estimation for details). Where possible, two
individuals of each species were sampled from widely spaced geographical locations to improve
the representation of intraspecific genetic variation.
8
Laboratory procedures
Genomic extraction methods were the same for both case studies. Standard salt-chloroform
procedures (Sambrook et al., 1989) were used to extract genomic DNA from ethanol preserved
muscle tissue of recently euthanized individuals. One nuclear and three mitochondrial gene
fragments for each case study were amplified using polymerase chain reaction (PCR) (Table
S2.1 and S2.2, Appendix A). Partial sequences of the cytochrome c oxidase subunit 1 (CO1),
12S rRNA and 16S rRNA genes were obtained using previously published PCR and sequencing
procedures and conditions (Read et al., 2006). The first intron of the S7 gene was amplified
using the primers S7RPEX1F (5′-TGG CCT CTT CCT TGG CCG TC-3′) and S7RPEX2R (5′-
AAC TCG TCT GGC TTT TCG CC-3′) following Chow and Hazama (1998). Each 20 μl
reaction volume contained 2.5 mM Tris–Cl pH 8.7, 5 mM KCl and (NH4)2SO4, 1.5–4.5 mM
MgCl2, 200 μM each dNTP, 10 μM each primer, 0.75U of Taq polymerase (Qiagen) and 10 ng
template DNA. Thermocycling profiles were carried out under the following conditions: an
initial denaturation at 94°C for 2 min, 35 cycles each consisting of denaturation at 94°C for 45
s, annealing at 50°C for 45 s, and extension at 72°C for 1 min, and a final extension at 72°C for
10 min. Following quantification by electrophoresis on 1.5% agarose gel, PCR products were
purified by isopropanol precipitation and directly sequenced in forward and reverse directions
using dye terminator chemistry, then cleaned following manufacturer protocols (Applied
Biosystems). Labelled sequenced products were analysed on an automated ABI3730XL DNA
analyser (Macrogen, Seoul, South Korea).
Sequence data
For the Anampses case study (Chapter 3), all specimens were used as separate taxa for
parsimony, likelihood and Bayesian analysis to establish whether or not evolutionarily
significant units (ESUs) were equivalent to species designations. Consensus sequences (50%
strict) were computed for age estimation following evaluation of the phylogenetic results. For
9
the Pomacanthus case study (Chapter 4), consensus sequences were constructed and used for all
analyses. Sequence assembly, editing and alignment were completed in GENEIOUS PRO v5.1
(Biomatters, Auckland, New Zealand; available at: http://www.geneious.com/) using default
settings. All alignments were manually adjusted through the insertion or deletion of gaps and
trimmed in order to minimize missing data. Care was taken to ensure that both alignments
analysed in the Anampses case study, the alignment used for phylogenetic analysis and the
consensus alignment used for age estimation, contained the same information. No gaps were
present in CO1 sequences, which also lacked stop codons when translated. Sequences obtained
from both case studies are available from GenBank (Table S2.1 and S2.2, Appendix A).
Appropriate substitution models were identified for each alignment and partition in
JMODELTEST (Posada, 2008). The corrected Akaike information criterion (AICc) (Sugiura,
1978; Hurvich & Tsai, 1989) was used to select models for further analysis because it allows
sample size to be specified and is the preferred method for small alignments.
Phylogenetic analysis
All analyses were performed on a dual core processor MacBook Pro or via the
Cyberinfrastructure for Phylogenetic Research (CIPRES) project (Miller et al., 2010), a browser
that provides access to TeraGrid resources for phylogenetic tree inference. Phylogenetic
relationships were assessed based on concatenated alignments using parsimony, likelihood, and
Bayesian inference analyses for each case study. PAUP* v4.0b10 (Swofford, 2003) was used to
perform maximum parsimony (MP) analysis for which five independent heuristic searches were
executed using a tree bisection–reconnection (TBR) branch-swapping algorithm with 100
random sequence addition replicates. Analyses were run with and without weights assigned to
transversions relative to transitions (2:1 for the Anampses dataset; 5:1 for the Pomacanthus
dataset). The weighted stepmatrix was calculated based on estimates of substitution rates
obtained from JMODELTEST (Posada, 2008). Nonparametric bootstrapping was also
10
implemented, with 1000 bootstrap pseudo-replicates using heuristic searches and TBR branch-
swapping with two random sequence additions per pseudo-replicate, to determine the level of
support for each clade. For each case study, a majority-rule consensus tree was produced from
all topologies recovered from the bootstrap analysis. The final weighted MP tree was chosen
based on consistency with other phylogenetic analyses.
Maximum likelihood (ML) analysis, implemented in GARLI v1.0 (Zwickl, 2006), was used
to search for the tree topology, branch lengths and substitution model parameters with the
highest log-likelihood (–lnL) score based on the maximum likelihood criterion. For each case
study, 10 independent iterations of the analysis were performed using default settings for
algorithm parameters. Of the ten best trees obtained, the tree with the best –lnL score was
considered the best overall topology. Since the program simultaneously estimates substitution
model parameters, the analyses were run with and without the substitution model specified (as
per AICc). This served to further ensure the analyses were converging on similar solutions for
each respective genus and not arriving in a suboptimal area of tree space. The final trees for
each genus used in further analysis were chosen based on the best overall –lnL scores. Bootstrap
analysis was also performed with 100 bootstrap replicates and the best ML tree specified as the
starting tree. For both Anampses and Pomacanthus, a majority-rule consensus tree was
constructed in PAUP* to determine the level of support for the clades recovered.
Bayesian inference (BI) methods were implemented in MRBAYES v3.1.2 (Ronquist &
Huelsenbeck, 2003) using a partition mix model on TG, a large NSF Tera-Grid Resource
accessed via CIPRES. Partitions were assigned according to gene region and the general
substitution model structure was specified (as per AICc) for each partition. Parameters were
unlinked across partitions and each was allowed to evolve under different rates using a flat
Dirichlet prior. For the Anampses case study, posterior probabilities of clades were calculated
following four 50 million generation Markov Chain Monte Carlo (MCMC) analyses, each with
four chains sampling every 1000 generations. For the Pomacanthus case study, posterior
probabilities were calculated following two 10 million generation MCMC analyses, each with
11
four chains sampling every 500 generations. Convergence was assessed using AWTY
(Wilgenbusch et al., 2004) and TRACER v1.5 (Rambaut & Drummond, 2007). Upon
examination of the trace files, a conservative burn-in of 10% (5 million generations, or 5000
trees for Anampses; 1 million generations, or 2000 trees for Pomacanthus) was discarded from
each run and a 50% majority-rule consensus tree was computed for each respective genus using
the remaining sampled trees.
Age estimation
Age estimation was performed using the BEAST v1.6.1 package (Drummond & Rambaut,
2007). Substitution models were set accordingly for each gene partition. A strict molecular
clock was applied to the CO1 partition in the Anampses analysis and to the 16S partition in the
Pomacanthus analysis following preliminary tests, which indicated a strict molecular clock
could not be rejected because the marginal posterior distribution of the standard deviation of the
substitution rate included zero. All other gene partitions rejected a strict molecular clock and a
relaxed uncorrelated lognormal clock was employed (Drummond et al., 2006). The birth–death
process was specified as the tree prior to account for speciation and extinction (Gernhard,
2008). Starting trees were randomly generated for each run.
Time calibrations were made based on fossil evidence, specifically for Anampses, the
minimum fossil age of Eocoris bloti (Bannikov & Sorbini, 1990) was used to place an
exponential prior on the root node (50 Ma hard lower bound; 95 Ma soft 95% upper bound)
following Cowman et al. (2009). For Pomacanthus, the fossil ages of Avitoluvarus dianae,
Avitoluvarus mariannae, Kushlukia permira and Luvarus necopinatus (Bannikov & Tyler,
1995) were used to date the most recent common ancestor (MRCA) of Luvarus, Zanclus and
Acanthuridae (Acanthurus and Naso) using an exponential prior (55.8 Ma hard lower bound;
63.9 Ma soft 95% upper bound) following Near et al. (2012). Posterior samples from four
independent MCMC analyses, each with 10 million generations sampling every 1000th
12
generation, were assessed for convergence and appropriate burn-in using AWTY and TRACER
v1.5. Tree files were combined, after removal of 10% burn-in, using LOGCOMBINER v1.6.1
(Drummond & Rambaut, 2007) and a maximum clade credibility tree was constructed for each
genus using TREEANNOTATOR v1.6.1 (Drummond & Rambaut, 2007) to display median ages
and 95% highest posterior density (HPD) intervals (upper and lower) for each node.
2.2 Multi-Family Phylogeny
Data selection and handling
Sequence data for analyses used in Chapters 5–7 were obtained from GenBank for those
coral reef fish families with distribution maps available from the IUCN Red List spatial
database (IUCN, 2011). At the time of data collection this included genera belonging to the
families: Acanthuridae, Chaetodontidae, Labridae, Pomacanthidae, and Epinephelidae. Species
designations were based on the IUCN Red List (IUCN, 2011) and FishBase
(http://www.fishbase.org). Only loci with the most coverage across all families were
considered. Among the mitochondrial loci, these included 16S rRNA, 12S rRNA, CO1, and
cytochrome b. Nuclear loci included TMO-4C4 and S7 intron 1. Sequence coverage was also
assessed within each of the reef fish families to maximize taxon sampling within genera and
minimize missing sequence data within the alignment. Only those genera where a minimum of
70 % of constituent species had sequences available for any of the four mitochondrial loci, and
at least 50 % sequence coverage across all six loci were considered. This resulted in the
inclusion of 53 genera within four coral reef fish families: Chaetodontidae, Labridae,
Pomacanthidae, and Epinephelidae (Table S2.3, Appendix A). Additional sequences were also
included for taxa used to root the phylogeny (Opsanus pardus and Porichthys notatus) and to
provide additional nodes for fossil calibration (Tables S2.3 and S2.4, Appendix A).
Sequences were aligned in GENEIOUS PRO v6.1.2 (Biomatters, Auckland, New Zealand;
available at: http://www.geneious.com/) using default settings for each locus. All alignments
13
were manually adjusted through the insertion or deletion of gaps, and trimmed to minimize the
amount of missing data. PARTITIONFINDER v1.0.1 (Lanfear et al., 2012) was used to
simultaneously select an appropriate partitioning scheme for the concatenated alignment and the
best fitting models of molecular evolution for each locus based on the Bayesian Information
Criterion (BIC) (Schwarz, 1978).
Spatial data were compiled from the IUCN Red List spatial database for all species
included in the phylogenetic analysis (except Aprops bilinearis, Grammistes sexlineatus,
Niphon spinosus, and Zalanthias kelloggi, which had not been assessed by the IUCN Red List
at the time of sampling). Spatial analyses were conducted in GRASS (GRASS Development
Team, 2011) and QGIS (QGIS Development Team, 2012). The geographical range area of each
species was calculated as the total area of all constituent polygons in their respective
distribution maps (IUCN, 2011).
Phylogenetic analysis and age estimation
A time-calibrated phylogeny was constructed based on partitioned Bayesian analysis in
BEAST v1.7.5 (Drummond & Rambaut, 2007) using nine fossil calibrations (Table S2.4,
Appendix A). Partitioning and models of molecular evolution were specified according to those
identified by PARTITIONFINDER (Lanfear et al., 2012) (Table S2.5, Appendix A). Divergence
times were estimated under a relaxed uncorrelated lognormal clock model (Drummond et al.,
2006) following preliminary analyses that rejected a strict molecular clock for all partitions. The
birth–death process was specified as the tree prior to account for speciation and extinction
(Gernhard, 2008), and starting trees were randomly generated for each run. Temporal
calibrations were made based on fossil evidence used in previous phylogenetic studies of fishes
(Cowman et al., 2009; Near et al., 2012) (Table S2.4, Appendix A). Posterior samples from six
independent MCMC analyses, each with 40 million generations and sampling every
2000th generation, were assessed for convergence and appropriate burn-in using TRACER v1.5
14
(Rambaut & Drummond, 2007). Tree files were combined using LOGCOMBINER v1.6.1
(Drummond & Rambaut, 2007) following the removal of 20 % burn-in and re-sampling every
4000 states. A maximum clade credibility tree was constructed using TREEANNOTATOR v1.6.1
(Drummond & Rambaut, 2007) to display median ages and 95 % HPD intervals for each node.
15
Chapter 3: The role of peripheral endemism in species
diversification: evidence from the genus Anampses (Family:
Labridae) Published in Molecular Phylogenetics and Evolution 2012 62(2): 653–663
3.1 Introduction
One of the central topics in marine evolutionary biogeography focuses on understanding
and explaining the pattern of increased species richness in the area known as the Indo-
Australian Archipelago (IAA). This area is often referred to as a biodiversity hotspot because it
encompasses the world’s largest concentration of marine biodiversity (Reaka et al., 2008;
Renema et al., 2008; Bellwood et al., 2012). Numerous ecological factors including habitat
diversity, reef area and geographical location correlate with patterns of species richness in and
around the hotspot (Bellwood et al., 2005; Hoeksema, 2007). Though insightful, correlative
ecological factors are based primarily on present day species’ distributions and ecologies, and
therefore contribute little to our understanding of how the diversity was generated. Large-scale
patterns of biodiversity, such as that documented in the IAA hotspot, result from interactions
between current ecological and environmental conditions, as well as the biogeographical history
of the area. The study of evolutionary lineages in relation to their geographical distributions has
played a crucial role in elucidating patterns of historical biogeography (Lessios et al., 2001;
Rocha et al., 2007; Floeter et al., 2008; Riddle et al., 2008; Leray et al., 2010). Recent
analytical advances have permitted the incorporation of historical evidence in phylogenetic
reconstructions, which has enabled the key issue of timing of species origination to be
addressed through the construction of detailed chronograms (Alfaro et al., 2007; Hurley et al.,
2007; Azuma et al., 2008; Frey & Vermeij, 2008; Cowman et al., 2009; Kazancıoğlu et al.,
2009; Santini et al., 2009; Bellwood et al., 2010). As a result, we are now able to tackle novel
16
questions regarding the evolution of peripherally isolated endemic species and to explore the
role they may play in shaping the diversity of this distinctive region.
Endemism has long been a topic of interest among marine researchers. The conventional
biogeographical hypotheses put forth to explain the bulls-eye pattern of diversity concentrated
in the IAA, typically referred to as the ‘Centre-of’ models (Hoeksema, 2007; Bellwood et al.,
2012), make explicit predictions about the ages and geographical locations of endemic species.
Under the Centre of Origin model, young endemics should lie in the IAA, while under the
Centre of Overlap and particularly, the Centre of Accumulation models, they should be
peripheral to the IAA. Early contributions to the study of endemism in the marine realm equated
hotspots to major centres of endemism, largely based on patterns known from terrestrial
systems (Roberts et al., 2002). Further work demonstrated that the majority of corals and reef
fish found to occur within the IAA are not endemics (Hughes et al., 2002). It is now broadly
recognised that most coral reef associated endemism is concentrated along the periphery of the
hotspot (Bellwood & Wainwright, 2002; Hughes et al., 2002; Jones et al., 2002; Connolly et
al., 2003; Meyer et al., 2005; Allen, 2008; Briggs, 2009; Budd & Pandolfi, 2010; Hobbs & van
Herwerden, 2010). Yet, previous molecular work has provided no consistent pattern with regard
to the ages of peripheral endemic species in favour of any one ‘Centre of’ model over another.
Furthermore, it has been acknowledged that no single explanation is likely to account for the
presence of the IAA hotspot (Rosen, 1984; Palumbi, 1997; Bellwood & Wainwright, 2002;
Cowman & Bellwood, 2011).
By their very nature, peripherally isolated endemic species do not add to the diversity of
species found within the central hotspot. However, by examining the evolutionary relationships
and ages of peripheral endemics and their closest relatives we may better understand what
speciation mechanisms are implicated in generating coral reef fish biodiversity at range edges.
To do this, I have focused on a widespread tropical Indo-Pacific reef fish genus with a relatively
high proportion (42%) of peripherally isolated endemic or restricted range species, several of
which have shared distributions. The genus Anampses (family: Labridae) is a relatively small
17
group consisting of 12 extant species, all of which were sampled. There are few distinguishing
morphological and meristic features among Anampses species and previous attempts to recon-
struct their evolutionary history have relied almost exclusively on the highly varied colour
forms present in this sexually dichromatic group of fishes (Randall, 1972). Knowledge of the
evolutionary history of Anampses will enable us to identify sister relationships of endemic
species and estimate when they diverged. By combining this information with species
distributions I am able to consider the speciation processes responsible for generating coral reef
fish diversity at range edges. Temporal concordance among endemic species will be of
particular interest for those species with shared distributions and may identify certain times or
events that facilitated diversification of peripheral endemic species.
With such a high proportion of endemic species within Anampses, I am are afforded an
opportunity to consider the relative risk of extinction faced by vulnerable, isolated species.
Older endemic species that may have experienced range contraction and remain as relictual
isolates could face an increased risk of extinction compared to younger endemic species whose
range has yet to reach its full extent. One of only two documented marine fish extinctions has
already occurred within the genus Anampses, namely Anampses viridis (Dulvy, 2006). The
species’ small geographical range, which included only coastal waters of Mauritius, is widely
acknowledged to have contributed significantly to its fate (Hawkins et al., 2000; Dulvy et al.,
2003). Recently, Russell & Craig (2013) have shown that A. viridis had been taxonomically
confused and that it is actually the adult male (terminal phase) colour form and a junior
synonym of Anampses caeruleopunctatus. Nevertheless, by estimating the ages of endemic
species I am able to inform conservation efforts, particularly in areas susceptible to habitat
degradation and loss associated with increasing stresses on contemporary coral reefs.
This study aims to examine the genus Anampses with recently developed analytical tools to
provide an overview of their evolutionary history and a detailed chronological understanding of
peripherally isolated endemic species. The specific objectives of this study were:
18
1) To construct a phylogeny of the genus Anampses using maximum parsimony,
maximum likelihood and Bayesian inference, as well as a fossil calibrated
chronogram using BEAST (Drummond & Rambaut, 2007) to identify the age of the
genus and its constituent species; and
2) To identify possible models of diversification responsible for generating endemic
species at range edges; and
3) To explore concordant spatial and temporal patterns among endemic species within
the Anampses clade; and finally
4) To discuss the role of peripheral endemism in shaping the diversity of the IAA and
evaluate the probable threats endemic species face in relation to their age.
3.2 Materials and Methods
Methods of Data selection, Laboratory procedures, Sequence data, Phylogenetic analysis
and Age estimation for this data chapter follow those detailed in Chapter 2: General Methods,
section 2.1 Case Studies.
Biogeographical analysis
Geographical range and occurrence data were compiled for Anampses species (Choat et al.,
1988; Hughes et al., 2002; Kuiter, 2010). These data were used to classify species into three
biogeographical range categories: endemic, mid-range and widespread. An endemic is
traditionally defined as a taxon restricted, or peculiar, to a locality or biogeographical province.
Here, endemics were defined as species with distributions restricted to peripheral bio-
geographical regions including islands, archipelagos and continental margins, with geographical
ranges less than 5 × 106 km2. I acknowledge that this definition encompasses some species with
distributions that span multiple islands and/or archipelagos. Mid-range species were defined as
19
those with geographical ranges between 5 × 106 and 20 × 106 km2, while widespread species
were defined as those with ranges exceeding 20 × 106 km2. Geographical ranges of closely
related species were mapped and used to explore possible models of diversification responsible
for generating endemic species.
3.3 Results
Phylogenetic analysis
The combined data matrix consisted of 1666 bp and yielded a strong phylogenetic signal
with 500 parsimony informative characters (30.0%). The 12S mitochondrial rRNA fragment
had 415 bp, CO1 contributed a further 53 bp and intron 1 of the nuclear fragment S7
contributed 720 bp. Models selected using the corrected Akaike information criterion (AICc)
criterion varied among loci as summarised in Table 3.1. For species with more than one
available sequence there was a mean pairwise identity of 97.1%.
Table 3.1 Gene-specific models of evolution of Anampses fishes selected under the corrected Akaike
information criterion (AICc) in JMODELTEST. Single gene alignments were analysed separately and the
appropriate model applied for further partitioned analysis of the concatenated alignment. NST specifies the
number of substitution parameters.
Data Set AICc Model Invariable Sites Gamma Shape (γ) Nst 12s TPM3uf + I + G 0.439 0.342 6
CO1 TIM1 + I + G 0.633 1.565 6 S7 TPM3uf + G — 1.643 6
Complete TIM2 + I + G 0.265 0.491 6
The best maximum likelihood (ML) tree was selected based on the –lnL scores and resulted
from the analysis without model specification in which parameters were estimated from the data
during the analysis. This analysis consistently returned the same tree topology (–lnL =
20
−12,032.916) from all 10 independent runs, suggesting that the overall optimal tree was
recovered every time. Bayesian inference (BI) analysis was assessed for convergence by
plotting pairwise comparisons of node (split) frequencies of all four independent runs (Fig.
S3.1, Appendix B). This diagnostic is assessed by the strength of the correlation between two
runs and demonstrated moderate convergence for the analyses performed here. Cumulative
posterior probabilities for the 20 most variable nodes (splits) for each Markov Chain Monte
Carlo (MCMC) run were also examined and confirmed relatively stable clade frequencies
indicating that the subsamples of trees were describing similar distributions (Fig. S3.2,
Appendix B).
The phylogeny of the genus Anampses, represented by the best ML tree, was the most
well resolved phylogeny consistent with all other analyses and showed strong bootstrap support
[maximum parsimony (MP) and ML] and posterior probability support (BI) for the nodes (Fig.
3.1). All analyses confirmed the monophyly of Anampses with 100% support and placed
Macropharyngodon and Halichoeres in a clade sister to Anampses. Species delineations were
consistent with evolutionarily significant units (ESUs) based on the specimens and collection
locations. The five endemic species within Anampses were spread throughout the two most
speciose clades. Interestingly, those with shared geographical distributions (Anampses
chrysocephalus and Anampses cuvier in Hawaii; Anampses elegans and Anampses femininus in
the south Pacific) arose in separate clades.
Age estimation
Log files from four independent BEAST analyses showed high effective sample sizes
(posterior ESS values > 200 for four combined analyses), which indicated valid estimates based
on independent samples from the posterior distribution of the MCMC. Assessment of
convergence in TRACER and AWTY indicated that all analyses had converged. Pairwise
21
Figure 3.1 Inferred phylogeny of Anampses obtained by Bayesian inference (BI), maximum likelihood
(ML) and maximum parsimony (MP) analyses of three loci (CO1, 12S and S7) with multiple sequences
for each species (see Table S2.1 in Appendix A for locality information). The topology shows the best
ML tree with bootstrap support (> 50%) from MP and ML (1000 and 100 bootstrap replicates
respectively) and posterior probabilities from BI (consensus of 10,401 trees). Asterisks denote 100%
support for the node across all three analyses.
comparisons of node (split) frequencies among all four independent runs demonstrated
correlation (Fig. S3.3, Appendix B). Examination of the cumulative posterior probabilities of
the 15 most variable nodes (splits) from each of the four independent runs revealed stable clade
frequencies (Fig. S3.4, Appendix B).
The maximum clade credibility chronogram was compiled in TREEANNOTATOR from
36,004 post-burnin trees (36,004,000 generations) returned from the BEAST analyses. Ages are
presented as median node heights with the 95% highest posterior density (HPD) intervals
109
822
1431
1282
979
05
06
1490
940
1293
1291
1399
186/692
1186
956
974
82
1170
252
251
776
825
502/761
380/208
75
230/511
1373
1374
1138
1139
197/607
38/621
73
97
98/70100/99
95/100100/100
100/100100/9675/53
94/67
80/52
100/98
100/79
BI/MLMP
70100
7492
9797
53
74
97
*
100/68
94/
60/
99/
*
*
*
*
*
*
*
*
*
*
*
*
*
A. chrysocephalusA. melanurus
A. lineatus
A. meleagrides
A. elegans
A. twistii
A. neoguinaicus
A. caeruleopunctatus
A. lennardiA. cuvier
A. femininus
A. geographicus
M. bipartitus
M. cyanoguttatus
H. hortulanus
H. marginatus
S. melops
22
indicated by horizontal bars at each node. Strong posterior probabilities were obtained for all
nodes (Fig. 3.2).
The Anampses lineage dates back to the mid-Eocene when it shared a most recent common
ancestor (MRCA) with Macropharyngodon and Indo-West Pacific Halichoeres 46.2 Ma (95%
HPD: 33.1–64.0 Ma). Subsequent divergence within the genus began in the early Miocene 19.6
Ma (95% HPD: 12.3–28.8 Ma) when Anampses geographicus emerged as the earliest extant
species. Speciation continued throughout the Miocene and a substantial portion of species
diversification (42%) occurred during the Pliocene (2.6–5.3 Ma) and Pleistocene (0.01–2.6 Ma)
epochs.
Whilst all species of Anampses occur in the Indo-Pacific realm (Fig. 3.3), five of the 12
extant species have peripherally isolated distributions, while only four species, Anampses
meleagrides, Anampses twistii, Anampses caeruleopunctatus and A. geographicus are
widespread throughout the Indo-West Pacific. An apparent Indo-Pacific split occurred between
two species (Anampses lineatus, which is widespread throughout the Red Sea and Indian
Ocean, and Anampses melanurus, which is widespread throughout the West Pacific) at
approximately 3.6 Ma (95% HPD: 2.0–5.7 Ma). The two Hawaiian endemics, A. cuvier and A.
chrysocephalus revealed distinct divergence times, 9.6 Ma (95% HPD: 6.0–14.1 Ma) and 1.72
Ma (95% HPD: 0.9–2.9 Ma), respectively. In contrast, the two endemic species whose
distributions include Lord Howe Island, A. femininus and A. elegans, have similar ages with
overlapping 95% HPDs, 9.6 Ma (95% HPD: 6.0–14.1 Ma) and 11.2 Ma (95% HPD: 7.2–16.2
Ma), respectively. Anampses lennardi has a sympatric distribution with A. caeruleopunctatus
and is the youngest endemic species with an estimated age of 0.6 Ma (95% HPD: 0.2–1.1 Ma).
3.4 Discussion
This is the first study to present a comprehensive and fully resolved molecular phylogeny of
the genus Anampses. All analyses strongly supported the monophyly of Anampses, with 12
23
distinct species. The phylogeny identifies evolutionary relationships within the genus and
presents a foundation for the estimation of ages. The diversification of Anampses during the
early Miocene is consistent with published estimates for other, closely related Indo-Pacific reef
fish genera (Bernardi et al., 2004; Barber & Bellwood, 2005; Read et al., 2006). Endemic
Anampses species varied in their ages, ranging from 11.2 Ma (A. elegans) to 0.6 Ma (A.
lennardi), and in their evolutionary relationships. No universal pattern was found. Most extant
endemic species occur at the range edges of Indo-Pacific coral reefs; they have arisen
throughout the evolutionary history of the genus, and display temporal concordance for some
isolated locations, but not for others.
Phylogenetic structure
The presence of three main clades within a monophyletic Anampses was consistently
supported by all analytical methods. Most relationships among species were consistent across
methods, with only minor differences in the relationships among Clade 3 taxa (specifically, the
relative placement of A. meleagrides, A. elegans, A. twistii and Anampses neoguinaicus to each
other in the inferred phylogeny versus the chronogram). The sister relationship between
Anampses and the clade containing Macropharyngodon and Indo-West Pacific Halichoeres is
concordant with previous molecular studies of the Labridae (Westneat & Alfaro, 2005;
Cowman et al., 2009; Kazancıoğlu et al., 2009; Cowman & Bellwood, 2011).
Traditional morphology-based evolutionary hypotheses for Anampses suggested similar
relationships to those presented herein (Randall, 1972). Earlier taxonomic debate has
highlighted A. geographicus, and several authors have suggested that the genus be divided into
two sub-genera: Pseudoanampses, containing just A. geographicus, and Anampses containing
the remaining 11 species (Randall, 1972). This proposal was based primarily on differences in
lateral-line scale counts (Pseudoanampses with 48–50 lateral-line scales and Anampses with 26
or 27). The results presented herein provide partial support for this suggestion, identifying A.
24
25
Figure 3.2 A fossil-calibrated chronogram of Anampses. Ages are represented as median node heights of
the 95% highest posterior density (HPD) interval from a maximum clade credibility tree complied from
post-burnin topologies of four independent Bayesian Markov Chain Monte Carlo (MCMC) analyses (10
× 106 generations per run) implemented in BEAST. Asterisks indicate 100% posterior probability for the
node. Photos, by Randall (Unpublished), depict the initial phase (left panel) and terminal colour phase
(right panel). ° indicates possible intermediate phase; a comparable terminal phase photo of Anampses
melanurus was not available. Circles indicate relative biogeographical range size based on three separate
categories; red circles represent peripherally isolated endemic species with geographical ranges < 5 × 106
km2.
geographicus as an early diverging lineage, within a single monophyletic Anampses genus.
However, A. geographicus as a monotypic sub-genus may be paraphyletic and subgeneric
divisions appear to be unwarranted.
Age estimation and the origin of endemics
The Anampses lineage began diversifying in the early Miocene and underwent further
expansion during the Pliocene and Pleistocene, a time when other coral reef fish genera were
also diversifying (Bernardi et al., 2002; McCafferty et al., 2002; Streelman et al., 2002; Klanten
et al., 2004; Alfaro et al., 2007). Indeed, the age of the MRCA of Anampses (~19.6 Ma) is
similar to the ages of other, closely related coral reef fish genera with comparable sampling
including Macropharyngodon (Read et al., 2006), Thalassoma (Bernardi et al., 2004) and
Halichoeres (Barber & Bellwood, 2005; see Cowman & Bellwood, 2011 for an overview). For
many modern day reef fish families, a shift onto coral reef habitats following the K/T boundary
mass extinction event (65 Ma) was associated with an increase in species richness (Bellwood &
Wainwright, 2002; Alfaro et al., 2007; Bellwood et al., 2010) particularly during the Miocene
(Cowman & Bellwood, 2011). Once established on coral reefs, the Labridae underwent a wave
of innovation during the Miocene resulting in the origination of many specialized feeding
26
Figure 3.3 Maps showing the biogeographical distribution of Anampses species. Clade notations
correspond to those in Figure 3.2. Species’ geographical ranges and occurrences were compiled from
Hughes et al. (2002) and Kuiter (2010).
geographicus (Z)
femininus (Y)
lennardi (W)
cuvier (X)
caeruleopunctatus (V)
Clade 3
V W X Y Z
elegans (U)
neoguinaicus (T)
twistii (S)
meleagrides (R)Clade 2 Clade 1
Clade 3cUTSR
chrysocephalus (N)
melanurus (O)
lineatus (P)
Clade 1Clade 2Clade 3aClade 3b
N PO
(a) Clades 1 & 2
(b) Clades 3a & 3b
(c) Clade 3c
27
modes (Cowman et al., 2009). The expansion of Anampses in the early Miocene was no
exception and signified the emergence of another highly specialized feeding mode. The genus
characteristically possesses a single pair of broad projecting incisiform teeth at the front of the
jaws (Randall, 1972). They use these highly modified teeth to help suck invertebrate prey,
mainly crustaceans, from the epilithic algal matrix. This feeding mode appears to be unique
among labrids (Wainwright et al., 2004) and probably played a key role in the ecological
success of the genus.
Discerning the biogeographical history of a given group of organisms from a phylogenetic
hypothesis is difficult (Losos & Glor, 2003; Heads, 2009) and there are often multiple modes of
speciation that can be invoked to explain observed patterns of relatedness and present-day
distributions (Kirkendale & Meyer, 2004). The four main modes of speciation are: allopatry,
which occurs when the range of a species is split into isolated populations (Mayr, 1942);
peripatry, which is simply allopatry along the periphery of a species’ distribution (Mayr, 1963);
sympatry, which is driven by factors such as natural selection and competitive exclusion and
occurs without geographical isolation (Coyne & Orr, 2004); and parapatry, which occurs when
there is restricted gene flow between two diverging populations (Gavrilets, 2000). Evidence for
all four modes of speciation has been reported for marine organisms (Lessios et al., 2001;
Munday, 2004; Rocha et al., 2005; Bernardi et al., 2008; Leray et al., 2010; Malay & Paulay,
2010; Bird, 2011; Fitzpatrick et al., 2011; Ingram, 2011). The peripheral location and timing of
endemic species divergence within the Anampses clade revealed possible differences in the
underlying modes of speciation. Although it is difficult to separate the major modes based on
phylogenetic data, there are three models of diversification that may help us understand which
modes are operating. These three models are the: (a) ‘successive division’, (b) ‘successive
colonization’ and (c) ‘peripheral budding’ models (Fig. 3.4). The relationship between A.
lineatus, A. melanurus and A. chrysocephalus in Clade 3c appears to have arisen from
‘successive division’ or ‘successive colonization’, or some combination of the two. Both
models involve species’ divergences coinciding with a directional shift in their distributions.
28
However, in ‘successive division’, vicariance events underpin allopatric speciation, whereby a
series of divisions results in the fragmentation of a widespread distribution. In contrast,
peripatric or parapatric speciation modes are behind ‘successive colonization’, whereby
individuals from the ancestral population enter a new environment in an adjacent or isolated
location.
The ‘peripheral budding’ model is less obvious from observed ages and topological
relationships among species. Nevertheless, the species in Clade 2 appear to serve as an example
of this model whereby it appears that peripatric, parapatric or sympatric speciation from a
relatively widespread species has led to several peripheral isolates throughout the evolutionary
history of the lineage, with the most recent bud displaying a sister relationship to the
widespread species from which it diverged. Thus, the ages and topological relationships in the
chronogram may be deceiving because the widespread species will appear much younger than it
actually is due to the fact that it is genetically most similar to the most recently diverged bud.
Similar models of diversification have also been proposed for gastropods and hermit crabs
(Kirkendale & Meyer, 2004; Malay & Paulay, 2010), with support from phylogeographic data
reported for the parrotfish species, Scarus psitticus (Winters et al., 2010). It is important to note
that without additional information from the fossil record it is not possible to substantiate the
model. Still, the implication of the ‘peripheral budding’ model for A. caeruleopunctatus is that
rather than being ~600 Ka, it is more likely older than 9.6 Ma, having given rise to budded
species A. cuvier, A. femininus, and A. lennardi (9.6 Ma, 9.6 Ma and 600 Ka, respectively).
The alternative hypothesis would suggest that A. caeruleopunctatus has acquired a
geographical range spanning the entire Indo-Pacific in less than 600 thousand years. Pleistocene
sea-level changes may be invoked to explain such rapid range expansion. Under this scenario,
peripheral isolates characteristically occupy reef slopes along the edges of continental margins
or oceanic islands and suffer weak bottlenecks and isolation during stands of low sea level.
Their populations would persist through fluctuations in sea-level due to the flexible nature of
the reef slope habitat and its ability to track changes in sea-level. In contrast, lagoonal species
29
would suffer stronger bottlenecks during periods of low sea level with substantial loss of
habitat. Following sea-level rise, the opportunity would exist for lagoonal species to rapidly
expand their ranges due to the re-appearance of suitable habitat (sensu Fauvelot et al., 2003).
Anampses caeruleopunctatus generally occupies inshore, protected reefs and lagoons (Randall,
1972; Kuiter, 2010), and while its age (~600 Ka) does coincide with a period of sea-level rise
(Lisiecki & Raymo, 2005), this was only one of many fluctuations in sea-level during the
Pleistocene. Population genetic data of these three endemic species and their widespread
sister/putative ancestor are required to critically evaluate these alternate hypotheses.
Temporal discordance among co-occurring endemics
The age estimates revealed both similarities and differences in the ages of endemic species
with shared isolated distributions. Anampses cuvier is a relatively old endemic (~9.6 Ma) found
only in the Hawaiian Islands and Johnston Atoll. It shares part of this isolated distribution with
another endemic species, A. chrysocephalus, which emerged more recently, ~8 Ma later. The
Hawaiian Islands are part of the Hawaiian-Emperor seamount chain, which formed over the last
85 million years (Clague & Dalrymple, 1987; Rooney et al., 2008). The chain formed as a result
of volcanic outpourings onto the Pacific lithospheric plate as it moved in a northwesterly
direction over a geographically fixed hotspot in the asthenosphere (Wilson, 1963; Rotondo et
al., 1981). Thus, the islands are ordered linearly by age, with the oldest island in the northwest
and the youngest in the southeast (5.1–0.43 Ma, respectively); over time, these volcanoes
eroded to form coral atolls and seamounts (Fleischer et al., 1998). Given the seamount chain’s
extensive geological history, A. cuvier could have colonized islands or seamounts in the area
~9.6 Ma, and subsequently expanded its range to other reef habitat on surrounding islands and
atolls once it became available. As the island chain progressed this created more habitat, which
would have allowed A. cuvier to stretch its distribution further from older, now submerged
atolls as they moved northward, to the newly formed islands in the southeast.
30
31
Figure 3.4 A schematic diagram of three alternate models of diversification observed among Anampses
species, the: (a) ‘successive division’, (b) ‘successive colonization’, and (c) ‘peripheral budding’ models.
In the first, sequential allopatric speciation leads to an apparent directional shift in species’ distributions.
In the ‘successive colonization’ model peripatric or parapatric speciation also leads to a directional shift
in species divergence and, in this example, results in the same phylogenetic topology as (a). In the
‘peripheral budding’ hypothesis peripatric, parapatric or sympatric speciation leads to the divergence of
several peripheral isolates throughout evolutionary history. This results in the ancestral species ‘A’
appearing as one of the youngest species due to its genetic similarity with the most recently diverged bud
(species ‘D’).
Anampses cuvier is sister to A. femininus, a peripherally isolated species whose range is
restricted to the area between the southern Great Barrier Reef and Easter Island. Anampses
femininus probably arose as a peripheral bud from the ancestral lineage of A.
caeruleopunctatus. The data suggest that A. femininus subsequently gave rise to its own bud, A.
cuvier, via its eastern most populations. Similar phylogenetic relationships are known for other
coral reef fish genera, specifically the Dascylus trimaculatus species complex where Dascylus
albisella in Hawaii is sister to Dascylus strasburgi in the Marquesas, rather than to Dascylus
auripinnis in the geographically closer Line and Phoenix Islands (Leray et al., 2010).
Additional evidence for genetic links between Hawaii and Easter Island has been reported in the
sea urchin genus Diadema (Lessios et al., 2001). Diadema paucispinum, a species considered
endemic to Hawaii, actually extends as far south as Pitcairn and Easter Islands. Further
exploration of extralimital sister-group relationships in the context of evolutionary time may
provide more support for this route of Hawaiian colonization.
Anampses chrysocephalus is a young species (1.7 Ma) that occurs in partial sympatry with
A. cuvier. These two species have not only arisen at different times, but they have also arisen in
different clades. Thus, the results offer no evidence for the dispersal of species from older to
younger islands as in some terrestrial fauna (Wagner & Funk, 1995) and no support for
secondary endemism (Rotondo et al., 1981; Bird et al., 2011). Rather, there appears to have
32
been an independent colonization by A. chrysocephalus. Multiple species’ colonization of the
Hawaiian Archipelago has occurred in other coral reef fish genera including the butterflyfish
genus Chaetodon (Bellwood et al., 2010; Craig et al., 2010), and the wrasse genus Thalassoma
(Bernardi et al., 2004). Evidence from the fossil record of corals and gastropods also supports
continuing, sporadic immigration to the Hawaiian Islands (Kay & Palumbi, 1987). Anampses
chrysocephalus is sister to A. melanurus, which is distributed throughout the west Pacific. Thus,
A. chrysocephalus may have utilized any number of corridors to colonize Hawaii, including
dispersal from Easter Island.
Temporal concordance among co-occurring endemics
Another isolated distribution is shared by A. elegans and A. femininus. The former, A.
elegans (~11.2 Ma) is endemic to Lord Howe Island, with temporary pseudo-populations in
northern New South Wales and the island region of northern New Zealand (Choat et al., 1988).
Anampses femininus (~9.6 Ma) also occurs at Lord Howe Island, but has a larger distribution
than A. elegans, occurring as far east as Easter Island. While the estimated ages of these species
may seem disparate, the 95% HPDs surrounding them overlap considerably. This raises the
prospect of peripherally isolated species evolving in two separate clades, at roughly the same
time and possibly in the same isolated location. Although the more widespread distribution of
A. femininus suggests that dispersal and range expansion could have allowed this species to
colonize Lord Howe Island at any time throughout its ~10 million year history. When the ages
of these two species were compared to the age of Lord Howe Island, which formed during a
short interval in the late Miocene (6.4–6.9 Ma) (McDougall et al., 1981), there is a mismatch;
the lineages leading to the endemics apparently arose prior to the island.
At first it may be assumed that these species are relictual, particularly with the placement of
A. elegans as the first to diverge in its lineage. However, Lord Howe Island is an eroded shield
volcano, which, like the Hawaiian Islands, is part of a chain of volcanic seamounts that formed
33
due to the northward movement of the Australian plate over a stationary mantel hotspot during
the Oligocene and Miocene (Wellman, 1983; Quilty, 1993; Exon et al., 2004). Following their
formation, the volcanoes subsided and were fringed by reefs, some of which have persisted until
the present day (McDougall & Duncan, 1988; Exon et al., 2006; Przeslawski et al., 2011). Lord
Howe Island is an example of such a volcano that lies at the southern end of the chain.
The geological history of the region combined with the age estimates suggests that both A.
elegans and A. femininus, as peripherally isolated species, occupy part of an island chain, the
current configuration of which is the presence of Lord Howe Island. Furthermore, it appears
that the age of an island is not a reliable indicator of the ages of its inhabitants. The Lord Howe
seamount chain is an example of how islands come and go (Woodroffe et al., 2010) and the
presence of A. elegans and A. femininus as peripheral isolates is similar to the pattern of
peripheral isolation observed in Hawaii. The creation of new habitat in the Lord Howe region
due to tectonic uplift and volcanism would have allowed ancestral Anampses species to expand
their ranges or found new populations on emerging islands during intervals of spatially
extended tropical/subtropical conditions (Quilty, 1993). For A. elegans, Lord Howe Island is
now the only stable population base, for A. femininus it is merely one of many. Other studies of
marine organisms within and nearby the Lord Howe Region have emphasised the importance of
isolation and limited dispersal (Samadi et al., 2006; Castelin et al., 2010) as drivers of genetic
divergence among populations living in this patchy habitat. Overall, the results demonstrate that
the divergence of peripherally isolated restricted range or endemic species may be facilitated by
the presence of fragmented habitat provided by seamount and island chains.
The role and relative extinction risk of peripheral endemics
As is the case for most coral reef associated endemics, the majority of endemic Anampses
are peripherally isolated relative to the IAA hotspot. Only one of the five endemic species
studied here, A. lennardi, has a distribution that may be considered part of the IAA. While
34
endemic species can represent newly diverged species (neo-endemics), or relatively old,
relictual species that may have suffered significant range contraction, it seems that regardless of
their age, peripherally isolated endemic species appear to have limited ability for range
expansion and possibly limited ability to undergo speciation events. This is most likely a
consequence of their occurrence on the periphery, or along the edges of the biogeographical
domain. Expansion outward is most likely restricted by abiotic physical and environmental
factors, while expansion inward may be limited by biotic interactions such as competitive
exclusion and introgression (Brown & Lomolino, 1998).
However, recent phylogeographic studies have highlighted the potential importance of
peripheral island populations in exporting larvae inward toward the IAA (Lessios et al., 2001;
Drew & Barber, 2009; Eble et al., 2011; Fitzpatrick et al., 2011; Skillings et al., 2011). This
suggests that some peripheral endemics may export diversity. Determining the extent to which
peripheral endemics contribute to the biodiversity of species within the IAA will require further
species-level phylogenetic studies. Population genetic studies may also be useful to estimate
effective population sizes through time, provided coalescence and divergence times are
comparable (Bernardi & Lape, 2005; Bowen et al., 2006b; van Herwerden et al., 2006; Klanten
et al., 2007). Studies of this nature will likely reveal cryptic patterns of peripheral speciation
(Budd & Pandolfi, 2010; Gaither et al., 2010; Hobbs & van Herwerden, 2010; Leray et al.,
2010; Fitzpatrick et al., 2011) and provide more detail about the role of peripheral isolation at
range edges in the evolution of Indo-Pacific coral reef biodiversity.
Regardless of their role, peripheral endemics are no doubt valuable within their ranges and
face the possibility of being overlooked in terms of conservation initiatives simply because of
their isolated locations relative to the bulk of coral reef biodiversity. The consequences of
isolation may have already been realized by A. viridis, a species once thought to be endemic to
the island of Mauritius (Hawkins et al., 2000; Dulvy et al., 2003; but see Russell & Craig,
2013). Randall (1972) suggested that A. viridis might have been a close relative of A.
caeruleopunctatus. This seems likely in light of the ‘peripheral budding’ hypothesis. It may be
35
that peripheral speciation is quite common and results in more extinction than previously
realized. Peripheral endemic and restricted range species are more susceptible to extinction
threats as a result of their limited ranges and relatively small population sizes, particularly if
they are specialists (Hawkins et al., 2000; Munday, 2004). Of the endemics considered here, A.
elegans has the smallest range and the oldest estimated age. These factors, combined with the
fact that its distribution is a narrow margin on the interface between tropical and subtropical
environments, could put it at increased risk of extinction. It is no doubt vulnerable to any
number of local impacts, which could cause global loss. Hopefully more consideration will be
given to peripheral areas with high proportions of endemic species when designing future
conservation initiatives.
36
Chapter 4: Evolution of sympatric species: a case study of
the coral reef fish genus Pomacanthus (Pomacanthidae) Published in Journal of Biogeography 2013 40(9): 1676–1687
4.1 Introduction
The study of the geographical distribution of organisms has made fundamental
contributions to our understanding of evolutionary phenomena, including speciation. Recently,
focus has shifted away from attempts to label divergence events according to speciation mode
(i.e. allopatric or sympatric) and is moving towards identifying the processes and isolating
barriers affecting divergence (Fitzpatrick et al., 2008; Bird et al., 2012; Santini et al., 2012).
Before one can investigate these processes and isolating barriers, it is necessary to establish
biogeographical patterns in conjunction with robust evolutionary hypotheses (Coyne & Orr,
2004; Butlin et al., 2008; Mallet et al., 2009). This study therefore focuses on patterns of
eventual sympatry, rather than sympatric speciation as a process, and considers the multiple
modes of speciation that can lead to sympatry among species.
Age–range correlation (ARC) analysis has been used as a comparative method to quantify
the relative importance of allopatric versus sympatric speciation (Barraclough & Vogler, 2000).
The method involves assessing the degree of sympatry (contemporary range overlap) against
species (node) age (Fitzpatrick & Turelli, 2006). It assumes that the extent of overlap between
ranges is dependent on the geography of speciation, but becomes randomized over time through
independent range changes (Barraclough & Vogler, 2000; Fitzpatrick & Turelli, 2006). If the
mode of speciation within a group of taxa has been predominantly allopatric, then sister-species
should have limited range overlap, and range overlap should increase with time and become
more prevalent among distantly related sister clades. The opposite pattern presents if speciation
37
has been predominantly sympatric, with recently diverged sister-species exhibiting greater
range overlap than more distantly related sister clades.
ARC analysis has been questioned on the basis that substantial changes to species
geographical ranges can effectively eliminate the relationship between the geography of
speciation and contemporary geographical distributions (Barraclough & Vogler, 2000; Losos &
Glor, 2003), especially in highly mobile species (Phillimore et al., 2008). Despite this, ARC
analysis has been shown to be capable of detecting when the predominant mode of speciation is
sympatric (Phillimore et al., 2008). Further improvements have been made as a result of
simulation studies, including: independent comparisons to determine overlap between clades,
and Monte Carlo methods to test the relationship between phylogenetic relatedness and
geographical range overlap (Fitzpatrick & Turelli, 2006); and the proportion of species showing
zero or complete range overlap as a more reliable indicator of the geography of speciation
(Phillimore et al., 2008). All of these methods require a robust phylogeny within a temporal
framework (Fitzpatrick & Turelli, 2006) and generally work best when applied to geographical
relationships among sister-species (Phillimore et al., 2008; Fitzpatrick et al., 2009).
Despite these advances, few studies have used ARC analysis to investigate the geography
of speciation in the marine realm (Williams & Reid, 2004; Frey, 2010; Krug, 2011; Quenouille
et al., 2011). Most subsequently concluded that the majority of sister-species pairs have
allopatric distributions and that allopatric speciation is likely to have been the predominant
mode (but see Krug, 2011). Upon application of ARC analysis to several genera of coral reef
organisms, Quenouille et al. (2011) found that most sister-species with contemporary sympatric
ranges were no younger than 4 Myr. They concluded that the predominant mode of speciation
among these genera was allopatry followed by range expansion. By applying ARC analysis, I
can investigate whether the same patterns are more widely applicable to coral reef fishes by
assessing evidence from the genus Pomacanthus.
38
Pomacanthus is a relatively small genus, with 13 extant species. This enabled complete
sampling of extant taxa, which is essential when analysing relationships of sister-species
(Barraclough & Nee, 2001; Santini et al., 2012). Pomacanthus species occur throughout the
Indo-Pacific and Caribbean, with extensive range overlap among species (92% of species occur
in sympatry), facilitating meaningful analyses of the relationship between geography and
evolution. This is the first study to use ARC analysis to investigate the relationship between
geography and speciation for coral reef fishes in the context of a fossil-calibrated phylogeny.
Previous studies have extrapolated the temporal scale of phylogenetic hypotheses using a range
of methods, not all of which have been consistent (Quenouille et al., 2011). Caution must be
exercised when analyses rely heavily on a temporal component, to ensure that the methods used
to determine such a context are both reliable and comparable.
This is also one of the first studies to employ relatively accurate geographical range maps
that limit the extent to which geographical ranges of species may be overestimated.
Overestimating geographical ranges by including areas with unsuitable habitat can potentially
skew spatial analyses such as ARC. Species distribution maps were obtained from the recently
compiled IUCN database (IUCN, 2011). These maps were used in conjunction with a fossil-
calibrated phylogenetic hypothesis to establish a spatial and temporal framework within which I
investigated the role of geography in the evolution of Pomacanthus species. My specific
objectives were:
1) To determine the phylogenetic relationships among Pomacanthus species within a
temporal context; and
2) To test whether the degree of sympatry among sister taxa correlates with their age
using a range of methods and indices; and
3) To evaluate the relative contribution of the three main speciation modes in explaining
the production of sympatric distributions.
39
4.2 Materials and Methods
Methods of Data selection, Laboratory procedures, Sequence data, Phylogenetic analysis
and Age estimation for this data chapter follow those detailed in Chapter 2: General Methods,
section 2.1 Case Studies.
Biogeographical analysis
Spatial data were compiled for all Pomacanthus species (IUCN, 2011) and analysed in
GRASS (GRASS Development Team, 2011) to determine the degree of sympatry among sister
taxa using the definition of Barraclough & Vogler (2000). Several methods were applied to
investigate geographical patterns of speciation. The relationship between geography and species
age was examined through ARC analysis by plotting the degree of sympatry against node age
across all nodes within the genus. Geographical ranges of higher nodes in the phylogeny were
reconstructed by summing the ranges of all species subtending the node (Barraclough & Vogler,
2000). Arcsine-transformed degree of sympatry was regressed onto node age to estimate an
intercept and slope. A negative slope with an intercept > 0.5 is suggestive of sympatric
speciation, where recently diverged sister-species exhibit high levels of range overlap compared
with more distantly related sister clades (Barraclough & Vogler, 2000). In contrast, an intercept
of < 0.5 and a positive slope suggests allopatric speciation as the predominant mode, where
sister-species have low levels of range overlap and range overlap becomes more prevalent with
time among distantly related sister clades (Barraclough & Vogler, 2000).
Independent comparisons (equivalent to independent contrasts; see Felsenstein, 1985) were
also used in the form of nested averages of pairwise overlap between species as an alternative
method of ancestral range reconstruction. This method provides an estimate of the average
range overlap between species, or clades at each node since divergence. Average range overlap
was regressed onto estimated node age. The null hypothesis of no relationship between
divergence time and mean overlap was tested using 100,000 Monte Carlo simulations and a
40
Mantel test (Dietz, 1983) implemented in R (R Development Core Team, 2011). The same
pattern of high intercept and negative slope indicative of sympatric speciation described above
for ARC analysis also applies to the independent comparisons analysis (Fitzpatrick & Turelli,
2006). The proportion of sister-species pairs with zero or complete range overlap was also
calculated and compared to the simulations of Phillimore et al. (2008).
Finally, the degree of range-size symmetry at each node was calculated following
Barraclough & Vogler (2000), to determine whether peripatric speciation may have been a
dominant mode within the genus Pomacanthus. The degree of range-size symmetry is bounded
between zero and 0.5, with the latter representing sister clades with equal-sized ranges.
Intercepts below 0.25 suggest the possible importance of small ranges in speciation. Post-
speciation range movements determine the slope of the relationship between range-size
symmetry and node age, where range movements within a finite area will lead to older clades
occupying a greater proportion of the total area and range-size symmetry may increase with
node age.
4.3 Results
Phylogenetic analysis
The combined data matrix yielded a strong phylogenetic signal with 301 parsimony
informative characters (18.5%). The models selected using corrected Akaike information
criterion (AICc) varied among loci (Table 4.1). No lack of convergence among Bayesian
inference (BI) analyses was detected by AWTY (Figs S4.1 and S4.2, Appendix C).
The phylogeny of the genus Pomacanthus, represented by the best BI tree, was the most
well resolved phylogeny consistent with all other analyses, and showed strong bootstrap support
[maximum parsimony (MP) and maximum likelihood (ML)] and posterior probability support
(BI) for most nodes (Fig. S4.3, Appendix C). All analyses confirmed the monophyly of
41
Pomacanthus with nearly 100% support and placed Holacanthus, Pygoplites and
Chaetodontoplus in a clade sister to Pomacanthus (Fig. 4.1, Fig. S4.3, Appendix C). There is a
clear, well-supported split between west and east Tethyan Pomacanthus species, with the west
Tethyan species restricted to the Caribbean and East Pacific, and east Tethyan species
distributed throughout the Indo-Pacific and West Indian Ocean. A further split within the east
Tethyan clade separated the Indo-Pacific species from the West Indian Ocean clade, with the
exception of Pomacanthus semicirculatus. Pomacanthus semicirculatus does not appear as a
basal lineage within the West Indian Ocean clade, but rather as a crown species, albeit
ambivalently supported at the crown within an otherwise totally supported crown clade of five
sister-species. Finally, there is another split within the west Tethyan clade, which separates the
two Caribbean species from Pomacanthus zonipectus in the East Pacific.
Table 4.1 Models of evolution of Pomacanthus fishes selected by JMODELTEST under the corrected
Akaike information criterion (AICc). Single gene alignments were analysed separately and the
appropriate model applied for further partitioned analysis of the concatenated alignment. NST specifies the
number of substitution parameters.
Data Set AICc Model Invariable Sites Gamma Shape (γ) Nst Data Set 12S 350 TIM2ef + I + G 0.372 0.789 6 16S 556 TIM3 + I + G 0.394 0.383 6 S7 705 TVM + G — 2.164 6
Complete 1611 SYM + I + G 0.158 0.638 6
Timing of speciation
Log files from four independent BEAST analyses showed high effective sample sizes
(posterior ESS values > 200 for four combined analyses), which indicated valid estimates based
on independent samples from the posterior distribution of the Markov chain. Assessment of
convergence in AWTY indicated that all analyses had converged (Figs S4.4 and S4.5, Appendix
C). The maximum clade credibility chronogram was compiled from 36,004 post-burn-in trees
42
(36,004,000 generations). Ages are presented as median node heights with the 95% highest
posterior density (HPD) intervals indicated by horizontal bars at each node. Strong posterior
probabilities were obtained for the majority of nodes (Fig. 4.1).
The Pomacanthus lineage dates back to the mid-Eocene when it shared a most recent
common ancestor (MRCA) with Holacanthus and Pygoplites c. 40 Ma. Subsequent divergence
within the genus began in the late Oligocene, around 24.5 Ma (95% HPD: 17.3–32.8 Ma), when
species in the east Tethys split from those in the west Tethys. Diversification continued
throughout the Miocene, with all extant lineages present by 5 Ma (minimum bound of 95%
HPD: 2.6 Ma).
Biogeographical patterns of speciation
Although the geographical separation of east and west Tethyan species is maintained, most
species within each of the four clades have a high degree of sympatry (Figs 4.2 and 4.3). Two
of the three west Tethyan species have partially sympatric distributions (66.7%), as do all 10 of
east Tethyan species (100%) (Fig. 4.4). The ARC analysis revealed a negative relationship
between geographical range overlap and species age (r = –0.09) (Fig. 4.5a). However, the
resulting values were not statistically significant (n = 12; d.f. = 1, 10; F = 0.04; P = 0.84).
Furthermore, Spearman’s rank correlation coefficient showed that the relationship between
degree of sympatry and node age was not significant (n = 12; d.f. = 10; t = 0.14; P = 0.66).
Despite the non-significant results, the intercept was higher than 0.5 (0.81, +0.19/–0.15 SE
untransformed; note that standard errors are asymmetrical because linear regressions were
performed on arcsine-transformed values for both standard ARC and range-size symmetry
analyses).
The ARC analysis using independent comparisons also yielded a negative relationship
between average geographical range overlap and time since divergence, but showed a steeper
43
Figure 4.1 Fossil-calibrated chronogram of Pomacanthus. Ages are represented as median node heights
of the 95% highest posterior density (HPD) interval from a maximum clade credibility tree compiled
from post-burn-in topologies of four independent Bayesian Markov Chain Monte Carlo analyses (10
million generations per run), implemented in BEAST. Asterisks at nodes indicate 100% posterior
probability. Photographs were obtained from FishBase (http://www.fishbase.org/).
decline in range overlap with increasing time since divergence compared to the standard ARC
analysis (r = –0.38; Fig. 4.5b). Monte Carlo simulations of average overlap failed to reject the
null hypothesis of no relationship between divergence time and degree of sympatry (n = 12; r =
0.038; P = 0.25). The independent comparisons ARC analysis generated an intercept of 0.79,
slightly lower than that generated by the standard ARC analysis, but still greater than 0.5.
44
The majority of sister-species comparisons showed complete or high range overlap (>
0.85); only 20% of sister-species pairs showed zero range overlap. Furthermore, the ages of
sister-species with sympatric distributions span the evolutionary history of the genus, ranging
from 5 to 13.2 Ma (95% HPD: 2.6–18.7 Ma). The average age of sympatric sister-species (7.3
Ma; ± 3.9 SD) was similar to the age of the only sister-species pair found in allopatry (P.
rhomboides–P. maculosus; 5.8 Ma; 95% HPD: 3.1–8.8 Ma), and to the age of the only species
that does not occur in partial sympatry with any other Pomacanthus species, P. zonipectus (8.6
Ma; 95% HPD: 5–12.8 Ma).
Figure 4.2 Geographical distributions of Pomacanthus species in the west Tethyan clade. Species’
distributions were obtained from IUCN Red List maps (IUCN, 2011). The degree of sympatry is shown
above each node as a percentage, where the geographical range of higher clades was reconstructed by
combining the ranges of all constituent species such that all areas where at least one species is found were
included. The average percentage overlap is shown below each of the higher nodes, where independent
comparisons were used in the form of nested averages of pairwise overlap between species. Hatched
areas overlaid onto shaded regions indicate sympatry among sister-species. The map uses the Mollweide
projection.
0 1000km
P. arcuatus
P. paru
P. zonipectus
99.1%
0%
0%
45
No consistent pattern among sister-species pairs was recovered from the analysis of range-
size symmetry (Fig. 4.6). The intercept was lower than 0.25 (0.17, +0.16/–0.02 SE
untransformed) and range-size symmetry increased with node age (r = 0.12), although this was
not statistically significant (n = 12; d.f. = 1, 10; F = 6.4 × 10–4; P = 0.98). In addition,
Spearman’s rank correlation coefficient showed that the relationship between range-size
symmetry and node age was not significant (n = 12; d.f. = 10; t = –0.15; P = 0.88).
4.4 Discussion
The results reveal that Pomacanthus is a monophyletic genus, within which a number of
speciation modes are likely to have been operating throughout evolutionary history. This is the
first study to produce a robust temporal phylogenetic hypothesis for all species of Pomacanthus
fishes. The phylogenetic reconstruction is consistent with previously published hypotheses and
age estimates based on DNA evidence (Bellwood et al., 2004). Several splits within
Pomacanthus correspond to major historical vicariance events such as the Terminal Tethyan
Event (TTE), and the rise of the Isthmus of Panama (IOP). There are an exceptionally high
proportion of sympatric sister-species and clades whose divergence times span the evolutionary
history of the genus. I find evidence consistent with sympatric speciation, based on the
relationship between degree of sympatry and species (node) age, using several methods and
indices. The results of the analysis of range-size symmetry, in combination with the general
patterns of lineage division within the genus, suggest that peripatric speciation in the form of
peripheral budding (cf. Chapter 3) may have also contributed to the richness of sympatric
species, especially in the West Indian Ocean.
46
0 2000km
(a)
(b)
(c)
(d)
(e)
(f)
P. semicirculatus
P. chrysurus
P. rhomboides
P. maculosus
P. asfur
100.0%
51.6%
P. semicirculatus
P. chrysurus
P. rhomboides
P. maculosus
P. asfur
0%
P. semicirculatus
P. chrysurus
P. rhomboides
P. maculosus
P. asfur
98.1%
P. xanthometopon
P. sexstriatus
P. navarchus
99.9%
96.8%
P. xanthometopon
P. sexstriatus
P. navarchus
86.6%
P. annularis
P. imperator
95.2%
47
Figure 4.3 Geographical distributions of Pomacanthus species in the east Tethyan clade. Species’
distributions were obtained from IUCN Red List maps (IUCN, 2011). The degree of sympatry is shown
above each node as a percentage, where the geographical range of higher clades was reconstructed by
combining the ranges of all constituent species such that all areas where at least one species is found were
included. The average percentage overlap is shown below each of the higher nodes, where independent
comparisons were used in the form of nested averages of pairwise overlap between species. Hatched
areas overlaid onto shaded regions indicate sympatry among sister-species. Maps use the Mollweide
projection.
Allopatry
There are few well-known vicariance events that have led to simultaneous divergence and
allopatric speciation across many groups of marine organisms. One well-documented example
is the TTE, which involved the collision of the African and Eurasian continental plates and the
subsequent loss of shallow marine habitats and marked the end of the connection between the
tropical Indian and Atlantic oceans (Bellwood & Wainwright, 2002). This took place over
millions of years, with the final closure of the Tethyan seaway between 12 and 18 Ma
(Steininger & Rögl, 1984). The TTE has been associated with a division between Caribbean and
Indo-Pacific taxa for many groups of marine organisms, including coral reef fishes (Bellwood et
al., 2004; Bernardi et al., 2004; Bellwood et al., 2010). The split of the Pomacanthus lineage
from its sister taxa c. 40 Ma pre-dates the final closure of the Tethyan seaway, but remains
consistent with other generic divisions of coral reef fishes (Bellwood et al., 2004; Cowman &
Bellwood, 2011). The divergence of the west Tethyan Pomacanthus species from those in the
east Tethys c. 25 Ma suggests that barriers to dispersal and panmixia in the central Tethys may
have facilitated allopatric divergence prior to the final closure of the seaway. Although the 95%
HPD intervals surrounding the deeper age estimates within the Pomacanthus phylogeny span
approximately 15 Myr, the results support previous suggestions (Bellwood et al., 2004) that the
marine influences of this major biogeographical boundary considerably pre-dated the final
closure of the Tethyan seaway at 12–18 Ma.
48
Figure 4.4 Areas of species’ range overlap within the Pomacanthus genus. The key in the upper-right corner details the corresponding number of species that occupy each
coloured region. The map uses the Mollweide projection.
49
Another biogeographical event that physically divided marine populations was the rise of
the IOP. The rise began about 16 Ma, with the final closure separating the Atlantic Ocean and
Pacific Ocean basins at 3.1 Ma (Coates & Obando, 1996). This provides an absolute minimum
age for geminate species divergence (Coates et al., 1992), although many divisions among reef
associated taxa pre-date the final closure, as ecological and genetic barriers were in place much
earlier (reviewed by Lessios, 2008). Bellwood et al. (2004) discussed the rise of the IOP in
relation to their estimated divergence of P. paru/P. arcuatus from P. zonipectus at 19.9 Ma,
highlighting that this estimate is considerably older than the minimum age of 3.1 Ma. The age
of this split estimated herein (8.6 Ma; 95% HPD: 5–12.3 Ma) is much younger and closer to the
final closure, but still pre-dates it. The older age estimate in the previous study may have arisen
from limited taxon sampling (Milne, 2009), a lack of nuclear markers (Brandley et al., 2011),
and/or different age estimation methods (Hug & Roger, 2007). The age estimated herein falls
well within the period of historical vicariance associated with the rise of the IOP (Lessios,
2008), providing further independent geological support of the fossil-calibrated phylogenetic
hypothesis.
The consideration of the age estimates in conjunction with contemporary geographical
distributions provides evidence in favour of two key biogeographical events driving allopatric
divergence at least twice during the evolutionary history of Pomacanthus.
Sympatry
Most Pomacanthus species (92%) occur in partial sympatry with at least one other
congeneric species. The results revealed that 80% of Pomacanthus sister-species also display a
high proportion of sympatry. For all sympatric sister-species, range overlap was above 85%. In
comparison to other marine taxa, this is one of the highest proportions of sympatric sister-
species, with only one other taxon (Haemulon/Inermia) reported to have a greater proportion
(Fig. 4.7). Furthermore, to my knowledge, intercepts as high as those obtained from the ARC
50
analyses have not been reported before. Such a high proportion of sympatric distributions
among sister-species suggests that sympatric speciation may have played an important role in
the evolutionary history of the genus. Similarly, the low proportion of Pomacanthus sister-
species pairs that showed zero range overlap suggests that sympatric speciation could account
for 70–100% of speciation events within the genus. This estimate is based on a simulation study
where varied levels of sympatric speciation and parameters were used to investigate the
relationship between range overlap and species age (Phillimore et al., 2008). The study found
that even when all simulated speciation was sympatric, a considerable proportion
(approximately 20%) of sister-species still showed zero range overlap. Therefore, the large
proportion of recent splits that exhibit patterns consistent with models of sympatric speciation
suggest that this mode of speciation may be common within Pomacanthus.
Despite the extent of range overlap among sister-species, it is not possible to discount the
notion that allopatric speciation followed by range changes could have produced the same
results. As with most coral reef fishes, Pomacanthus species have great potential for dispersal.
Although highly site-attached as adults, they are broadcast spawners whose larvae spend
extended periods of time in the plankton (ranging from 17 to 24 days for Indo-Pacific species;
Thresher & Brothers, 1985) with the ability to swim considerable distances (Stobutzki &
Bellwood, 1997), which clearly demonstrates the potential for shifting ranges. Furthermore,
large-scale historical biogeographical range shifts among coral reef fishes and other marine
organisms are well documented (Renema et al., 2008; Cowman & Bellwood, 2013a).
Contemporary geographical ranges can, however, be informative with respect to the past if
range shifts occur in such a way that spatial relationships are conserved (Fitzpatrick & Turelli,
2006).
Evidence from mammals, given their more complete fossil record, suggests the relative
position of contemporary geographical ranges may provide some information about the
geographical mode of speciation, even though ranges have shifted. Specifically, Fitzpatrick &
Turelli (2006) found that 97% of the 62 co-occurring congeneric species they examined in the
51
Figure 4.5 Plots of the degree of sympatry versus node age in the Pomacanthus phylogeny. Grey bars
represent the 95% highest posterior densities of node ages. Open circles denote sister-species pairs.
Diagrams to the right of each plot represent the gradient of range overlap along the y-axis. (a) Filled
circles mark nodes of higher clades where geographical ranges were obtained by summing the area where
at least one constituent species is found. (b) Filled circles mark nodes of higher clades where independent
comparisons, or nested averages of pairwise overlap between species in each clade were used as an
alternative to the union method above.
fossil record also show range overlap today. If there has been long-term maintenance of
sympatry within Pomacanthus, this would explain the non-significant relationship between
range overlap and species age. Unfortunately, it is not possible to substantiate whether or not
spatial relationships have been conserved through time for Pomacanthus species, or coral reef
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15 20 25 30
Node age (Ma)
Deg
ree
of s
ympa
try
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15 20 25 30
Node age (Ma)
Aver
age
% o
verla
p
(a)
(b)
35
35
52
fishes in general, due to the scarcity of appropriate fossils. However, all the evidence presented
herein suggests that sympatry is a common and sustained feature in pomacanthids.
I found no species, sympatric or otherwise, substantially younger than 4 Myr, with the
youngest species pair – P. paru and P. arcuatus – estimated to have diverged 5 Ma (95% HPD:
2.6–8.1 Ma). Indeed, there are few species pairs of coral reef fishes younger than 4 Ma (but see
Bellwood et al., 2010; Cowman & Bellwood, 2011; Choat et al., 2012 for exceptions), making
it potentially difficult to quantify the amount of time required for sister-species to acquire
sympatric distributions. Furthermore, in contrast to the findings of Quenouille et al. (2011), no
sharp transition from sister-species in allopatry to those in sympatry was detected in the more
recent evolutionary history of Pomacanthus. The results do not, however, disagree with the
suggestion made by Malay & Paulay (2010) in their study of speciation in hermit crabs, that a
substantial amount of time (> 2 Ma) may be required for sister-species to develop sympatric
distributions following allopatric divergence.
Figure 4.6 Plot of the degree of symmetry versus node age in the Pomacanthus phylogeny. Grey bars
represent the 95% highest posterior densities of node ages. Open circles denote sister-species pairs.
Diagrams to the right of each plot represent the gradient of symmetry along the y-axis. Filled circles mark
nodes of higher clades where geographical ranges were obtained by summing the area where at least one
constituent species is found.
0.0
0.1
0.2
0.3
0.4
0.5
Node age (Ma)
Deg
ree
of s
ymm
etry
0 5 10 15 20 25 30 35
53
Ecological knowledge can often inform the relative likelihood of sympatric speciation
within a group of taxa (reviewed by Rocha & Bowen, 2008; Puebla, 2009). The ecological data
currently available for Pomacanthus species provide little information on how processes of
sympatric speciation may have taken place, although more detailed data may yet reveal
ecological differentiation.
Overall, the best evidence obtained for sympatric speciation was the high proportion of
sister-species pairs with substantially overlapping contemporary geographical distributions.
Although this could be due to secondary contact as a result of shifting or expanding ranges, or
simply the increased survival of taxa whose ranges encompass the IAA and thus accumulate in
the IAA (Cowman & Bellwood, 2013a), the extent of overlap among sister-species pairs
detected herein is comparable with models of high levels of sympatric speciation. The results
therefore lend support to the possible importance of sympatric speciation in the evolutionary
history of Pomacanthus.
Peripatry
The intercept and slope of the analysis of range-size symmetry are consistent with
simulations of peripatric speciation within a finite area (Barraclough & Vogler, 2000),
suggesting that peripatry may also be an important mode of speciation within Pomacanthus.
This is particularly apparent when focusing on the topological relationships among species in
the West Indian Ocean clade. Pomacanthus semicirculatus is the only widespread species with
a distribution spanning the Indo-Pacific; all of the other Indian Ocean species have fairly
restricted distributions centred at varying locations along the east coast of Africa. Peripatric
speciation may have dominated in this African clade in the form of peripheral budding (sensu
Chapter 3). Under this scenario, a relatively widespread P. semicirculatus would have given rise
to several peripherally isolated species through peripatry, parapatry or sympatry. Maintaining
the order of lineage divergence observed within the clade, these species would have been
54
successively isolated along the east coast of Africa (with initial isolation of Pomacanthus asfur,
then P. rhomboides/P. maculosus, and later P. chrysurus). If this occurred primarily through
founder events, subsequent range expansion must have led to contemporary sympatry among all
constituent species.
Evidence for peripheral budding has been reported for other marine organisms (Pandolfi,
1992; Kirkendale & Meyer, 2004; Budd & Pandolfi, 2010; Malay & Paulay, 2010; Winters et
al., 2010; Chapter 3) and is likely to be more common than previously recognized.
Unfortunately, it can confound many of the inferences made from phylogenetic hypotheses,
including topological relationships among species and age estimation (Chapter 3). Therefore,
the consideration of all possible speciation scenarios when interpreting phylogenetic hypotheses
is advised, especially when the hypotheses are then applied to other questions and used in
analyses such as ARC.
Figure 4.7 Proportion of sympatric sister-species among marine taxa for which there exists a species-
level phylogeny with near complete taxon sampling (> 85%) and more than two sister-species pairs.
Sympatry is defined here as > 10% overlap. Numbers above bars indicate the total number of sister-
species pairs within each taxon. Phylogenies were obtained from the following studies: Chlorurus, Choat
et al. (2012); Thalassoma/Gomphosus, Bernardi et al. (2004); Nerita, Frey (2010); Dascyllus,
McCafferty et al. (2002); Echinolittorina, Williams & Reid (2004); Anisotremus, Bernardi et al. (2008);
Scarus, Choat et al. (2012); Anampses, Chapter 3; Mesoplodon, Dalebout et al. (2008); Naso, Klanten et
al. (2004); Pomacanthus, this study; Haemulon/Inermia, Rocha et al. (2008).
0.0
0.2
0.4
0.6
0.8
1.0
Chlorur
us
Thalas
soma/G
omph
osus
Nerita
Dascy
llus
Anisotr
emus
Scarus
Anamps
es
Mesop
lodon
Pomac
anthu
s
Haemulo
n/Ine
rmia
Pro
porti
on s
ympa
tric
sist
er s
peci
es p
airs
69 17 4
316
4
45
8
Naso
7
Echino
littori
na
16
55
4.5 Conclusions
This study explored potential patterns of speciation among all 13 Pomacanthus species by
comparing extant geographical distributions in the context of a fossil-calibrated phylogenetic
hypothesis. The lack of a direct relationship between the degree of sympatry and node age can
be explained by the likelihood of multiple speciation modes operating throughout the
evolutionary history of this genus. In order for a significant relationship between range overlap
and node age to be detected, there must have been a single predominant geographical mode of
speciation in the data set (Fitzpatrick & Turelli, 2006). Furthermore, if sympatric speciation has
been the dominant mode and Pomacanthus species have maintained sympatric distributions
over long periods of time, this would also explain the lack of a relationship between range
overlap and node age. Nevertheless, I found support for all three major modes of speciation and
evidence of a lengthy and evolutionarily complex history in pomacanthid fishes.
56
Chapter 5: On the relationship between species age and
geographical range in reef fishes: are widespread species
older than they seem? In press Global Ecology and Biogeography (2014)
5.1 Introduction
Identifying the evolutionary dynamics of species’ geographical ranges is a central goal of
macroecological studies. Three processes shape geographical ranges: speciation, range-size
transformation (which can be influenced by factors such as dispersal ability and habitat
connectivity), and extinction (Gaston, 1998). The resultant relationship between the age of a
species and its geographical range depends on the probability that species with different range
sizes will go extinct or speciate, and the way in which ancestral geographical ranges are
partitioned between daughter species (Webb & Gaston, 2000). Four basic theoretical models
describe how geographical ranges may change through time. The ‘age and area’ model predicts
constant geographical range expansion over time, producing larger ranges with increased
evolutionary age (Willis, 1922; Hubbell, 2001). The ‘stasis’ or ‘post-expansion stasis’ model
states that species achieve their maximum geographical ranges rapidly and maintain a uniform
trajectory thereafter (Jablonski, 1987; Gaston, 1998; Gaston, 2003). Both of these models
describe instantaneous range crashes corresponding to the geological timing of species’
extinctions.
In contrast, the ‘taxon cycle’ model predicts that species will expand their initially restricted
ranges to become maximally dispersed for a period of time until ranges begin to decline
towards extinction or the beginning of a new cycle (Wilson, 1961; Ricklefs & Cox, 1972;
Ricklefs & Bermingham, 2002). The ‘taxon cycle’ model can produce symmetrical (cf. Foote,
57
2007) or asymmetrical trajectories (cf. Webb & Gaston, 2000) depending upon the rate of range
expansion relative to range decline. All three of the aforementioned models predict initial
geographical range expansion, but discriminating between them is difficult because factors such
as extinction and sampling bias can influence the likelihood of detecting newly diverged or
near-extinct species and their respective geographical ranges (Pigot et al., 2012). Biologically,
the ‘taxon cycle’ model is arguably the most logical because species with smaller distributions
are generally at greater risk of extinction (Jablonski & Hunt, 2006). However, there need not be
a general pattern in the evolution of species’ geographical ranges through time. A fourth
theoretical model, the ‘idiosyncratic’ model proposes that species exhibit individual trajectories,
producing limited similarity in the range sizes of closely related species and the potential for no
over-arching pattern in range size evolution through time among a given group of taxa (Gaston,
2003).
Studies from a wide variety of groups have shown a positive relationship between species
age and geographical range size including, aquatic beetles (Abellan & Ribera, 2011), bats
(Weber et al., 2014), birds (Webb & Gaston, 2000; Böhning-Gaese et al., 2006), frogs
(Wollenberg et al., 2011), minnows (Taylor & Gotelli, 1994), marine molluscs (Jablonski,
1987; Miller, 1997), and tropical plants (Paul & Tonsor, 2008; Paul et al., 2009). However, the
relationship between species age and range size is often variable and can be clade specific
(Webb & Gaston, 2000). Studies on large, diverse taxonomic groups such as New World bird
species (Gaston & Blackburn, 1997), mammals (Jones et al., 2005), and coral reef fishes (Mora
et al., 2012) found no consistent relationship between species age and geographical range.
To examine how species age relates to geographical range it is essential to have accurate
estimates of both parameters. While there has been much recent success in phylogenetic age
estimation techniques, conceptual synthesis of the inherent problems with these techniques has
been limited. For example, the use of divergence time estimates, obtained from the nodes of a
temporally-calibrated phylogeny (or relative node position e.g. Taylor & Gotelli, 1994), as an
indication of species’ ages is common, but beyond examining uncertainty in the tree or
58
calibrations, few other issues have been examined. Here I address the effects of phylogenetic
age estimation on the relationship between species age and geographical range. I argue that
estimates of divergence time are not always indicative of species age, and if considered as such,
they may mask any underlying relationship between species age and geographical range. I
highlight an example of this using coral reef fishes and explore the insights these methods bring
to light for models of reef fish geographical range evolution.
5.2 Phylogenetic Methods of Age Estimation
Zuckerkandl and Pauling (1965) first proposed the idea of dating evolutionary divergences
using calibrated differences among proteins. Their idea was based on evidence that sequence
differences between proteins among species varied as a function of the time since their
evolutionary separation. This central idea gave way to the development of molecular clock
methods for time-calibrating phylogenetic hypotheses (reviewed by Kumar, 2005). Since their
development, phylogenetic methods of age estimation have been increasingly applied to a
myriad of studies as a means of investigating the general mechanisms and processes of
evolution. In general, these methods involve converting the pairwise percentage of sequence
divergence between two species (or clades) into an estimate of the time since the most recent
divergence (Nei, 1987), hereafter referred to as divergence time. This provides a divergence
time estimate (relative or absolute, depending on the methods used to calibrate the tree) for
every node in the phylogeny, which represents the most recent common ancestor (MRCA)
between two taxa. To examine the relationship between species age and geographical range,
divergence times for all nodes in the phylogeny subtended by extant species are typically
considered to reflect species age (cf. Taylor & Gotelli, 1994; Gaston & Blackburn, 1997; Webb
& Gaston, 2000; Böhning-Gaese et al., 2006; Paul & Tonsor, 2008; Paul et al., 2009; Abellan
& Ribera, 2011; Mora et al., 2012; Weber et al., 2014). Therefore, divergence times that are
shared between two subtending sister-species (defined by their close genetic relationship) are
assigned to both daughter species and plotted against each of their corresponding geographical
59
ranges (Fig. 5.1a, b). To illustrate how extinction and certain models of diversification can bias
the potential relationship between species age and geographical range using these methods, and
how to account for the potential biases, I consider an example using coral reef fishes.
5.3 Quantifying Species Age in Reef Fishes
The traditional phylogenetic approach
A comprehensive genetic dataset of coral reef fish genera with near-complete sampling of
described species was constructed and combined with available fossil data to simultaneously
infer the phylogeny and estimate divergence times of species from four of the major coral reef
fish families. Sequence data were obtained from GenBank for those coral reef fish families with
distribution maps available from the IUCN Red List spatial database (IUCN, 2011). Molecular
data were assessed and selected to maximize coverage across all fish families, to maximize
taxon sampling within genera, and to minimize the amount of missing sequence data in the
alignment (see Chapter 2: General Methods, section 2.2 Multi-Family Phylogeny for
specific selection criteria and potential limitations). This resulted in the inclusion of four
mitochondrial loci (16S rRNA, 12S rRNA, CO1, cytochrome b), and two nuclear loci (TMO-
4C4, S7 intron 1) for 53 genera within four coral reef fish families: Chaetodontidae, Labridae,
Pomacanthidae, and Epinephelidae (Table S2.3, Appendix A). Sequences were also included
for additional taxa to root the phylogeny and provide nodes for fossil calibration. A time-
calibrated phylogeny was constructed based on six independent partitioned Bayesian analyses
in BEAST v1.7.5 (Drummond & Rambaut, 2007) using nine fossil calibrations (Table S2.4,
Appendix A). Log files showed high effective sample sizes (posterior ESS values > 200 for six
combined analyses), which indicated valid estimates based on independent samples from the
posterior distribution of the Markov chain. The maximum clade credibility chronogram was
compiled from 48,003 post-burn-in trees (48,003,000 generations) (Table S5.1 and Fig. S5.1,
Appendix D). Biogeographical data were also compiled and used to quantify species’
60
Figure 5.1 (a) Methods for time-calibrating nodes from a phylogeny using molecular data involve
converting the pairwise percentage of sequence divergence between two species (A and B) into an
estimate of the time since the most recent divergence. This is done the same way for deeper nodes in the
phylogeny, except the mean of the pairwise divergence between A and C, and between B and C is used.
These methods provide a divergence time estimate (x2 and x1) for every node in the phylogeny, which
represents the most recent common ancestor between two taxa. (b) The relationship between species age
and geographical range is typically examined by considering the estimated divergence times for all nodes
in the phylogeny subtended by extant species as representative of species age. Thus, divergence times
that are shared among two subtending sister-species are assigned to both daughter species and plotted
against each of their corresponding geographical ranges as (x2, yA), (x2, yB) and (x1, yC).
geographical range areas following the general methods outlined in Chapter 2, section 2.2
Multi-Family Phylogeny.
First, divergence time estimates for all nodes in the phylogeny subtended by extant species
were considered as an indication of species age (as described in Fig. 5.1a, b). I tested the
theoretical ‘age-area’ model of geographical range evolution using linear regression and found
no significant linear relationship between divergence time and geographical range for species
included in the phylogenetic analysis (r 2 = 0.00789; n = 293; d.f. = 1, 291; F = 2.31; P = 0.129)
(Fig. 5.2). The ‘stasis/stasis post-expansion’ and ‘taxon cycle’ models predict species’ range
expansion from initially restricted ranges in a curvilinear manner. To test these curvilinear
theoretical models I fitted a quadratic model to our reef fish data, but found no evidence for a
cyclical form of range evolution (r 2 = 0.00942; n = 293; d.f. = 2, 290; F = 1.38; P = 0.254). The
x2
A
B
yA
yB
(a) (b)
C yC
x1
y =
geog
raph
ical
rang
e
x = divergence time
x2 x1
yB
yA
yC
61
majority of reef fish species sampled were estimated to have diverged ~3.3 Ma (mean = 5.4 Ma)
during the Pliocene. The oldest extant lineage was 59 Ma.
Phylogenetic methods are only able to estimate the age of the MRCA between two species
when data for both species are present in the analysis. Therefore, taxa missing from the
phylogeny, whether extinct or extant, will cause the ages of sister taxa to be over-estimated
(Fig. 5.3). If the rate of extinction through time has remained relatively constant, then older
lineages are more susceptible to extinction (Rabosky & Lovette, 2008). Furthermore, the
potential degree of divergence time overestimation increases with branch length (i.e. the degree
of divergence time overestimation is potentially greater for species D compared with species B
in Fig. 5.3, given extinct species z and x, because species D has a longer branch and therefore a
greater range of time across which it could have shared a common ancestor with species z). It is
worth noting that if an extinct (or missing extant) species is not sister to a single extant species
Figure 5.2 Relationship between species age and geographical range plotted as described in Figure 5.1,
where the estimated divergence times for all nodes in the phylogeny subtended by extant species are
considered as species’ ages. This analysis included 293 reef fish ingroup species from four families:
Chaetodontidae, Labridae, Pomacanthidae, and Epinephelidae (Table S2.3, Appendix A). Mean estimated
species age was 5.4 Ma; modal species age was 3.3 Ma; and the range in species age was 58.5 Ma.
05
1015
20
Divergence time (Ma)
0 10 20 30 40 50 60
25
Geo
grap
hica
l ran
ge a
rea
(106 km
2 )
62
Figure 5.3 Effects of extinction, or missing extant taxa, on phylogenetic estimates of divergence time.
The letters A, B, C, and D subtending lineages with solid lines denote sampled extant species. Extinct, or
missing extant species, are denoted by x, y, and z subtending dashed lineages. Excluding species from the
phylogeny, whether extinct or extant, will cause the ages of sister-species to be overestimated. For
example, species x is sister to B, but missing from the phylogeny. Therefore, the divergence time
estimated for species B (t3) is older than it would otherwise be if x had been included. Assuming a
relatively constant rate of extinction, older lineages are more susceptible to effects of extinction.
Furthermore, older lineages have greater potential range for overestimation of divergence times compared
to more recent lineages, simply because they have longer branches. For example, there is a greater range
of time for missing species z to have diverged from D then there is for missing species x to have diverged
from B. Thus, it is likely that the estimated divergence time obtained for species D (t1) will be more
greatly overestimated than that of B (t3), given the existence of extinct or un-sampled species. If an
extinct or missing species is not sister to a single extant species, as in missing species y, this results in a
missing data point when examining the relationship between divergence time and geographical range,
rather than an overestimation of divergence time.
t3
A
B
C
D
t2
t1
x
y
z
TTime
t0
63
this results in a missing data point when examining the relationship between divergence time
and geographical range (e.g. extinct species y in Fig. 5.3), rather than an overestimation of
divergence time.
The impacts of extinction on the estimation of divergence times have been reviewed
previously (Harvey et al., 1994; Ricklefs, 2007), and approaches to minimize these impacts
have provided valuable insights into processes associated with divergence, such as rates of
phenotypic change (Seddon et al., 2013). One way to avoid these problems is by focusing on
sister-species. When robust sampling of extant species is achieved, sister-species relationships
can be reconstructed with confidence. Focusing on recent divergences between pairs of sister-
species minimizes the impact of extinct taxa on divergence time estimation, although it may not
be eliminated entirely. This technique narrows the focus of the question to address how
geographical ranges have evolved among the most recently diverged species. The impact of
missing extant taxa on estimates of divergence time will thus depend on the shape of the
phylogeny and sampling bias; and a sister-species approach will eliminate some of the impact
that missing extant taxa may have on estimates of divergence time.
A sister-species approach
Using the phylogeny of reef fishes, I considered divergence time estimates for all sister-
species pairs, excluding monotypic genera, as indicators of species age. This approach
minimizes the effects missing extant taxa may have on species age estimation, but additional
steps are required to prevent the pseudoreplication of divergence time estimates among sister-
species (as in Fig. 5.1a, b). To overcome this issue I first took the average of each sister-pair’s
geographical range and compared it with their estimated divergence time. Average geographical
range increased linearly with increasing species age (r 2 = 0.0481; n = 91; d.f. = 1, 89; F = 4.50;
P = 0.0368) (Fig. 5.4). The relationship can also be described using a curvilinear model (r 2 =
0.0713; n = 91; d.f. = 2, 88; F = 3.38; P = 0.0386). However, model comparison using ANOVA
64
showed that the curvilinear model was not a significant improvement upon the linear model (n
= 91; d.f. = 2, 88; F = 2.20; P = 0.142).
Another way to overcome the pseudoreplication of divergence time estimates among sister-
species is to randomly assign one of two potential geographical ranges to each sister-species’
divergence time estimate. The advantages of this approach are that it eliminates the problem of
pseudoreplication of divergence time estimates, and it selects geographical range data from only
one of the two sister-species without making any prior assumptions of the models under which
species have diversified. In addition, when the randomisation approach is repeated, it can
illustrate the likelihood, or percentage, of a significant relationship. I performed 1000 bootstrap
replicates of geographical range randomisation; for each replicate I fitted linear and curvilinear
models to the data. A significant linear relationship between species age and geographical range
was detected in 13% of the randomised replicates (the range of statistical values across all 130
significant replicates: r 2 = 0.111–0.0426; n = 91; d.f. = 1, 89; F = 11.12–3.96; P = 0.00125–
0.0497; plot not shown, see the vertical grey bars in Fig. 5.4 for the differences in sister-species’
geographical ranges).
The relationship between species age and geographical range could also be described by a
curvilinear model in 29% of the randomised replicates (the range of statistical values across all
290 significant replicates: r 2 = 0.157–0.0495; n = 91; d.f. = 2, 88; F = 8.17–2.29; P =
0.000885–0.0499). However, model comparison performed using ANOVA revealed that that
the curvilinear model was only a significant improvement over the linear model 4.8% of the
time (the range of statistical values across all 48 significant replicates: n = 91; d.f. = 2, 88; F =
7.94–3.96; P = 0.00598–0.0498). These applications of a sister-species approach help to
overcome some of the problems associated with estimating species’ ages from phylogenetic
hypotheses and they have unveiled the potential for a positive relationship between species age
and geographical range in reef fishes.
65
Figure 5.4 Minimum divergence times are used as indicators of species age for all sister-species pairs,
excluding monotypic genera, and plotted against the pairs’ mean geographical range area. Grey vertical
bars show the difference in geographical range areas between sister-pairs. Mean geographical range area
increases linearly with increasing species age (r 2 = 0.0481; n = 91; d.f. = 1, 89; F = 4.50; P = 0.0368).
The relationship can also be described by a curvilinear model (r 2 = 0.0713; n = 91; d.f. = 2, 88; F = 3.38;
P = 0.0386); however this model was not a significant improvement upon the linear model following
model comparison in ANOVA (n = 91; d.f. = 2, 88; F = 2.20; P = 0.142).
However, even when utilizing a sister-species approach it is important to consider the
spatial component of speciation when evaluating divergence time estimates in a phylogenetic
hypothesis, as certain models of diversification can confound them. For instance, under some
models of peripatric speciation such as peripheral budding (Chapter 3), a centrally distributed,
widespread species successively gives rise to multiple species along the periphery of its
geographical range (Fig. 5.5a). As a result, topologically the widespread species will appear to
be sister to the most recently diverged (budded) species and its age will thus be grossly
underestimated (Fig. 5.5b). Evidence for this mode of diversification has been reported for
marine (Pandolfi, 1992; Kirkendale & Meyer, 2004; Budd & Pandolfi, 2010; Malay & Paulay,
2010) and terrestrial taxa (Hoskin et al., 2011; Garrigan et al., 2012; Toussaint et al., 2013). If
this is a common occurrence in the diversification of coral reef fishes, as increasing evidence
05
1015
20
Minimum divergence time (Ma)
25
Mea
n ge
ogra
phic
al ra
nge
(106 km
2 )
0 2 4 6 8 10 12 14
66
suggests (Winters et al., 2010; Fitzpatrick et al., 2011; Choat et al., 2012; Chapters 3, 4), it
would mask any relationship between species age and geographical range when analysed using
age estimates for all extant species in the phylogeny. It may also mask the relationship when
analysed using a sister-species approach, depending on the methods used to assign geographical
ranges to species age estimates. If we consider this potential bias in the light of the initial results
of this relationship for reef fishes, it may help explain the presence of numerous species with
relatively young ages and large geographical ranges (Fig. 5.2). These large-range species may
appear to be young only because they have recently budded a peripheral species. They may be
much older.
Alternatively, young species may be able to attain large ranges given the considerable
potential for larval dispersal of most coral reef fish species (Strathmann et al., 2002). For
example, young species with restricted distributions would have the potential to expand their
geographical ranges into areas of suitable habitat over several successive generations. However,
many recent studies suggest that the level of self-recruitment among coral reef fish populations
is much higher than expected given this dispersal potential (Swearer et al., 2002; Bowen et al.,
2006a; Priest et al., 2012). Thus, although fishes have great dispersal potential it may not be
consistently realized and dispersal alone may not be substantial enough to decouple any
prospective relationship between species age and geographical range area.
Comparing minimum divergence time with minimum geographical range
To overcome the potential problem of underestimating the divergence times of widespread
species that have successively given rise to multiple species through time, I suggest using a
sister-species approach and considering the estimated minimum divergence time as an
indication of species age only for the species with the smaller of the two geographical ranges.
This method therefore examines the relationship between the minimum divergence time
(species age) and the minimum geographical range area. This approach also eliminates the
67
Figure 5.5 (a) Illustrates the peripheral budding model (sensu Chapter 3), also known as ancestral
persistence. A centrally distributed, widespread species E successively gives rise to multiple species
along the periphery of its geographical range from time t0 to time t3. (b) The topological relationships
resolved from a molecular phylogeny of species that have diverged under the peripheral budding model.
Lineage dashing corresponds to range outline in part (a). Species E and H are sister because they
diverged most recently and are molecularly most similar. However, the successive peripheral splitting, or
budding, of species causes the divergence time for species E to be grossly underestimated as t3 rather than
t0. Thus masking any potential relationship between species age and geographical range when analysed
using the methods described in Figure 5.1. This can be overcome by using a sister-species approach and
considering the estimated minimum divergence time as indicative of species age only for the species with
the smaller of the two distributions (e.g. plot t3 against the geographical range size of species H). This
method examines the relationship between minimum divergence time and minimum geographical range
area. This approach also eliminates the problem of pseudoreplication of species’ ages where a single
(shared) age is plotted against multiple geographical range areas (as in Fig. 5.1b).
problem of pseudoreplication of species’ ages where a single (shared) age is plotted against
multiple geographical range areas (as in Fig. 5.1a, b).
I compared species age with minimum geographical range area to illustrate the potential
impact successive peripheral speciation, or peripheral budding, may have on the relationship
t1
t2
(a)
E
F
t3
G
Ht3
t2
t1
E
F
G
H
(b)
TTime
t0
68
between these two variables for reef fishes. A significant positive linear relationship between
species age and minimum geographical range areas was found using this approach (r 2 = 0.104;
n = 91; d.f. = 1, 89; F = 10.35; P < 0.01) (Fig. 5.6). While the model still has fairly low
predictive power, it is one of the strongest linear relationships obtained overall (only one out of
the 1000 randomised replicates had a higher r 2 value). Furthermore, the difference in the trend
recovered after excluding potential biases associated with ancestral persistence suggests that
ancestral persistence may be prevalent among coral reef fishes, with successive peripheral
speciation events impacting area-age relationships.
Minimum geographical range area also increased in a curvilinear fashion with increasing
species age (r 2 = 0.124; n = 91; d.f. = 2, 88; F = 6.25; P = 0.00289; Fig. 5.6). However, the
results of model comparison using ANOVA showed that the curvilinear model was not a
significant improvement upon the linear model (n = 91; d.f. = 2, 88; F = 2.03; P = 0.157). The
task of discriminating between theoretical models of range evolution remains difficult;
however, the results for reef fishes suggest that the ‘age-area’ model may sufficiently describe
the early range expansion dynamics of species.
Because both variables, geographical range area and species’ age, were positively skewed
all of the preceding analyses were repeated on log-transformed data. A significant log-log linear
relationship would suggest that as species age, their geographical ranges increase exponentially.
Such a pattern may be consistent with the ‘stasis post-expansion’ or ‘taxon cycle’ theoretical
models of geographical range evolution if range expansion occurs rapidly. No significant linear
relationships were detected between the transformed variables in the preceding analyses, except
for the randomisation approach, for which there was a significant log-log relationship between
geographical range and species age in 9.3% of replicates (the range of statistical values across
all 93 significant replicates: r 2 = 0.0993–0.0427; n = 91; d.f. = 1, 89; F = 9.82–3.97; P =
0.00234–0.0495) (Fig. S5.2c, Appendix D). The statistical results of all of the log-transformed
analyses are reported and discussed in Appendix D.
69
Figure 5.6 A sister-species approach where minimum divergence time (as an indication of species age) is
plotted against minimum geographical range area (i.e. the smaller of two geographical range areas for all
sister-species pairs, excluding monotypic genera). As minimum divergence time increases, minimum
geographical range area also tends to increase. The relationship can be described by a linear model (r 2 =
0.104; n = 91; d.f. = 1, 89; F = 10.35; P = 0.00181). While the correlation is not strong, it is the strongest
linear relationship I obtained overall. The relationship can also be described by a curvilinear model (r 2 =
0.124; n = 91; d.f. = 2, 88; F = 6.25; P = 0.00289); however, this model was not a significant
improvement upon the linear model according to our ANOVA model comparison (n = 91; d.f. = 2, 88; F
= 2.03; P = 0.157). Because both variables, geographical range area and species’ age, were positively
skewed I repeated all of the preceding analyses on log-transformed data. The statistical results of the log-
transformed analyses are reported in Appendix D.
Species as sampling units are non-independent because they share an evolutionary history.
Therefore, the relationship between species age and geographical range recovered from the
sister-species approaches could be an artefact of the phylogenetic relationships among species.
Phylogenetically independent contrasts (PICs) were used to evaluate the consistency of this
relationship when accounting for the phylogenetic topology (Felsenstein, 1985). For variables,
species age and geographical range area, I optimised an OU model of character evolution and
performed linear regression analyses using PICs in R (R Development Core Team, 2011;
packages “geiger” and “ape”). Only the approaches that recovered previous significant results
05
1015
2025
0 2 4 6 8 10 12 14
Minimum divergence time (Ma)
Min
imum
geo
grap
hica
l ran
gear
ea (1
06 km2 )
70
were tested. PICs for the randomisation approach were calculated using the average
geographical range across all 1000 randomisation replicates. For this analysis, averaging across
bootstrapped replicates approximates the sister-species approach where geographical ranges
were averaged for each sister-pair. Thus, the two approaches produced similar results. When
tested for consistency using PICs the significant relationships between species age and
geographical range were not recovered (mean geographical range: r 2 = 0.0252; n = 91; d.f. = 1,
89; F = 2.30; P = 0.133; randomised replicates: r 2 = 0.0245; n = 91; d.f. = 1, 89; F = 2.24; P =
0.138; Fig. 5.7a, b, respectively). Therefore, phylogenetic structure may explain the original
relationship recovered from these two sister-species approaches (although it may also reflect a
loss of power as a result of the PIC approach). In contrast, the relationship between species age
and geographical range was consistent following the PIC analysis of the sister-species approach
comparing species age with minimum geographical range (r 2 = 0.0599; n = 91; d.f. = 1, 89; F =
5.67; P = 0.0194; Fig. 5.8). Thus in the context of phylogenetic relationships, only the
minimum geographical range area between two sister-species was significantly related with
species age in our reef fish example. It is likely that detailed information on geographical range
dynamics was lost through the randomisation or averaging of geographical ranges in the other
two sister-species approaches.
Geographical range symmetry
To further explore the dynamics of geographical range evolution among reef fishes I
investigated whether species age also effects the distribution of range size symmetry among
sister-pairs. I have shown that as species age increases, on average, the minimum geographical
range between sister-pairs also increases. Given this relationship, changes in range symmetry
with species age can be described using three possible scenarios of range dynamics. Firstly, if
the rate of range expansion were greater for the smaller-ranged species among sister-pairs, then
range size symmetry would increase, on average, with increasing species age. Range
71
Figure 5.7 The results of phylogenetically independent contrasts (PICs), which were performed to
evaluate the consistency of the relationship between species age and geographical range when accounting
for the phylogenetic topology (Felsenstein, 1985). An OU model of character evolution was optimised for
both variables and linear regression analyses using PICs were conducted in R (R Development Core
Team, 2011; packages “geiger” and “ape”). Only approaches that recovered previous significant results
were tested. (a) The PIC analysis of the average geographical range area between sister-pairs showed a
non-significant relationship with species age (r 2 = 0.0252; n = 91; d.f. = 1, 89; F = 2.30; P = 0.133). (b)
Similarly, the sister-species approach where geographical ranges between sister-pairs were randomised
and averaged across 1000 bootstrap replicates recovered a non-significant result when the phylogeny was
considered (r 2 = 0.0245; n = 91; d.f. = 1, 89; F = 2.24; P = 0.138).
contraction of the large-ranged species could also produce an increase in range symmetry with
species age. Secondly, if both species in a sister-pair expand their ranges at the same rate, then
minimum geographical range size would still increase with species age, but range symmetry
would remain constant. Finally, if the rate of range expansion were slower for the smaller-
ranged species, then range symmetry would decrease with species age. Substantial range
contraction of the large-ranged species could also produce a decrease in range symmetry with
species age.
Mea
n ge
ogra
phic
al ra
nge
cont
rast
( 106 )
Minimum divergence timecontrast
(a) (b)
-21
01
-1 0 1 2 3
23
-20
24
Minimum divergence timecontrast
-1 0 1 2 3
Geo
grap
hica
l ran
ge
cont
rast
( 106 )
72
Any of the three aforementioned scenarios of range dynamics may result if peripheral
budding has occurred. However, peripheral speciation events would generally result in sister-
pairs with asymmetrical ranges, thus one may expect asymmetry to be more prevalent among
younger sister-pairs if peripheral speciation is common in reef fishes. However, the occurrence
of young, asymmetric sister-pairs does not require diversification to have proceeded under the
peripheral budding model, it may instead reflect single, rather than successive, peripheral
speciation events.
The degree of geographical range symmetry among sister-pairs was incorporated into the
linear model describing the relationship between species age and minimum geographical range
area using multivariate analysis of variance (MANOVA) and Pillai’s test. Symmetry was
calculated following Barraclough & Vogler (2000), where the smallest geographical range area
between two sister-species was divided by the sum of their geographical range areas. Symmetry
values range from zero to 0.5 and higher numbers indicating greater symmetry.
Figure 5.8 The relationship between species age and geographical range was consistent following the
PIC analysis when I tested the sister-species approach comparing species age with minimum
geographical range (r 2 = 0.0599; n = 91; d.f. = 1, 89; F = 5.67; P = 0.0194).
Min
imum
geo
grap
hica
lra
nge
cont
rast
( 106 )
Minimum divergence timecontrast
-20
24
-1 0 1 2 3
73
As species age increases, the minimum geographical range between sister-pairs increases
and their ranges become more symmetrical (n = 91; d.f. = 2, 88; Pillai’s statistic = 0.116; P =
0.00442). To visualise the relationship between species age and both dependent variables, I
divided range symmetry into three categories, asymmetrical species were those with symmetry
values ≤ 0.1, symmetrical species were those with symmetry values ≥ 0.4, and those species
with symmetry values between 0.1 and 0.4 were defined as having medium symmetry.
Symmetry categories were added to the plot of species age and minimum geographical range
(Fig. 5.9). Over half of the asymmetrical species (53.8%, n = 13) diverged less than 2 Ma,
compared to 44.4% of those with medium symmetry (n = 54) and 37.5% of symmetrical species
(n = 24). Furthermore, I found a significant difference in the mean ages of asymmetrical (mean
= 2.0 Ma) and symmetrical (mean = 3.7 Ma) sister-species using a Welch two-sample t-test (n =
36; d.f. = 34.0; t = 2.18; P = 0.0367) (Figure 5.9). This result persisted when symmetry
extremes were defined at 10% (graph not shown; asymmetric mean species age = 1.9 Ma;
symmetric mean species age = 3.7 Ma; n = 21; d.f. = 14.2, t = 2.2; P = 0.0429).
I performed univariate tests of linear (r 2 = 0.0688; n = 91; d.f. = 1, 89; F = 6.58; P =
0.0120; Fig. 5.10) and curvilinear models (r 2 = 0.0690; n = 91; d.f. = 2, 88; F = 3.26; P =
0.0431; Fig. 5.10) to visualise the relationship between species age and the degree of
geographical range symmetry. The results of model comparison using ANOVA showed the
curvilinear model was not a significant improvement over the linear model (n = 91; d.f. = 2, 88;
F = 0.0123; P = 0.912).
The patterns of geographical range symmetry observed among reef fishes support
peripheral speciation as a significant factor in the generation of reef fish biodiversity. The
results suggest that as species age, they expand their geographical ranges in a way that produces
greater symmetry between the geographical ranges of sister-pairs.
74
Figure 5.9 Incorporates geographical range symmetry into the model relating minimum geographical
range to species age. Geographical range symmetry was calculated following Barraclough and Vogler
(2000), where the smallest geographical range area between two sister-species was divided by the sum of
their geographical range areas. Symmetry values range from zero to 0.5 and higher numbers indicate
greater symmetry. Symmetry was divided into three categories: asymmetrical sister-species with
symmetry values ≤ 0.1 (black circles), symmetrical sister-species with symmetry values ≥ 0.4 (white
circles), and sister-species with symmetry values between 0.1 and 0.4 were defined as having medium
symmetry (grey circles). As species age increases, the minimum geographical range between sister-pairs
increases and their ranges become more symmetrical (n = 91; d.f. = 2, 88; Pillai’s statistic = 0.116; P =
0.00442). Over half of the asymmetrical species (53.8%, n = 13) diverged less than 2 Ma, compared to
44.4% of those with medium symmetry (n = 54) and 37.5% of symmetrical species (n = 24).
Furthermore, a significant difference in the mean ages of asymmetrical (mean = 2.0 Ma) and symmetrical
(mean = 3.7 Ma) sister-species pairs was detected using a Welch two-sample t-test (n = 36; d.f. = 34.0; t
= 2.18; P = 0.0367).
5.4 Applications and Implications
The method I present herein narrows the question of species’ geographical range evolution
to the early dynamics of expansion. Because the method focuses on recent divergences between
pairs of sister-species, the detection of range decline or crash preceding extinction is less likely.
Nevertheless, our sister-species approach of comparing minimum divergence time (as an
indication of species’ age) with the minimum geographical range area between sister-pairs is
05
1015
2025
Min
imum
geo
grap
hica
l ran
ge ( 1
06 km2 )
0 2 4 6 8 10 12 14
Minimum divergence time (Ma)
75
Figure 5.10 The relationship between species age and the degree of geographical range symmetry can be
described by linear (r 2 = 0.0688; n = 91; d.f. = 1, 89; F = 6.58; P = 0.0120) and curvilinear models (r 2 =
0.0690; n = 91; d.f. = 2, 88; F = 3.26; P = 0.0431); however, the results of model comparison using
ANOVA showed the curvilinear model was not a significant improvement over the linear model (n = 91;
d.f. = 2, 88; F = 0.0123; P = 0.912).
applicable to any diversification scenario – it need not be used only in cases where peripheral
budding or ancestral persistence is suspected. Even if species diverged in allopatry, our method
simply re-phrases the question to explore the minimum geographical range area attainable by
species given their age. It provides a logical means of removing the pseudoreplication of
species’ ages and does not require the use of averaging geographical range areas.
I advocate the use of this method as a tool to explore the dynamics of range evolution and
note that its use in comparison with other approaches can potentially reveal interesting patterns.
The example I presented using reef fishes was based on contemporary geographical range data.
I acknowledge that the results may change given future additions of geographical range data or
nominal species. All of the methods examined in the present study assume that range size
transformations occur over periods of time concordant with the time-scale of the analyses.
Relatively short-term, idiosyncratic, cyclical range transformations may confound the results of
any phylogenetic analysis of geographical range dynamics. Comparative phylogeographic
studies may be better suited to detect these range transformations over shorter time-scales.
0.0
0.1
0.2
0.3
0.4
0.5
Geo
grap
hica
l ran
ge s
ymm
etry
0 2 4 6 8 10 12 14
Minimum divergence time (Ma)
76
However, successive peripheral isolation of populations or lineages within species can produce
the same issues with divergence time estimation as those discussed herein. Thus, our sister-
species method can be adapted into a sister-lineage method when examining range dynamics
within-species or populations.
Overall, the observations on how phylogenetic age estimation can affect the relationship
between species age and geographical range size will have important implications in
macroecology and biogeography. The methods I present herein provide a new way of exploring
the evolutionary dynamics of species’ geographical ranges. By establishing methods to
minimize the impacts of extinction and successive peripheral speciation on divergence time
estimation, I hope to stimulate further investigation into the prevalence of this model of
diversification across a broad range of taxa.
77
Chapter 6: Temporal evolution of coral reef fishes: global
patterns and disparity in isolated locations Published in Journal of Biogeography 2014 41(11): 2115–2127
6.1 Introduction
Coral reef fishes are exceptionally diverse throughout the world’s tropical oceans. To
understand the foundation of such high biodiversity, and to help predict the response of
contemporary patterns of diversity to future environmental change, we study the distribution of
organisms through space and time. Over the past 150 years, many authors have described
spatial patterns of marine diversity (reviewed by Bellwood et al., 2012; Briggs & Bowen, 2012;
Kulbicki et al., 2013). However, the molecular techniques and analytical tools to assess
distributional patterns through evolutionary time have only been developed recently. These
tools allow us to examine the impacts of differing geological histories on the age structure of
fauna from distinct biogeographical areas, or within isolated locations.
Spatial patterns of diversity are generally described by the delineation of areas with unique
biotas. Biogeographical divisions of marine environments have been based on the concentration
of endemic species (Ekman, 1953; Briggs, 1974; Briggs & Bowen, 2012), shared biotic and
environmental characteristics (Longhurst, 1998; Spalding et al., 2007), or quantitative
assessments of community composition (Floeter et al., 2008; Kulbicki et al., 2013). These
methods have produced varying degrees of division and identified areas that often differ in
boundary placement and scale. However, most workers have recognised the importance of
historical isolation in separating two major biogeographical realms: the Indo-Pacific and the
Atlantic. Each realm has experienced a distinctive history that has shaped its constituent fauna –
from ongoing speciation as part of the expanding biodiversity hotspot that has developed in the
78
central Indo-Pacific since the Miocene (Renema et al., 2008; Cowman & Bellwood, 2013a), to
substantial loss of many marine taxa in the Atlantic during a period of faunal turnover in the
Plio-Pleistocene (Bellwood, 1997; O'Dea et al., 2007). Given such different evolutionary
histories, one may expect to find distinct realm-specific temporal patterns in the evolution of
coral reef fishes. My first aim was to investigate these realm-specific patterns.
With increased taxon sampling and increasingly congruent molecular sequence data, I am
also able to evaluate the predictions of the four main models of diversification (Centre of
Origin, Overlap, Accumulation or Survival) used to explain the hotspot of biodiversity in the
central Indo-Pacific [referred to as the Indo-Australian Archipelago (IAA) or Coral Triangle]
(the ‘Centre of’ hypotheses are reviewed briefly in Palumbi, 1997; Barber & Bellwood, 2005;
and in detail in Bellwood et al., 2012). There are two contrasting scenarios. If the IAA were a
Centre of Accumulation or Overlap of species ranges (Gaither & Rocha, 2013), one would
expect to find younger species in peripheral locations outside the IAA. These hypotheses
suggest that species arise in isolated, peripheral locations (such as the island arcs in the West
Pacific) and remain there as endemics until they disperse to the IAA. Conversely, if the IAA
were the Centre of Origin, one would expect to find the youngest species within this region. The
Centre of Survival hypothesis suggests that species can arise anywhere, but they survive better
in the IAA.
If species arise at the same rate across all biogeographical regions, then there should be no
pattern in the distribution of young endemics. But if species arise in proportion to regional
species richness, then most endemics would be located in the IAA. Consequently, the
geographical distribution of young endemics can help evaluate these alternate hypotheses. My
second aim therefore was to examine the extent of peripheral speciation as a mode of
diversification in reef fishes. To specifically examine peripheral locations, I compared the ages
of endemic species from two disparate locations with high coral reef fish endemism, the Red
Sea and the Hawaiian Islands (Allen, 2008). If either location has been a significant area of
79
recent species origination, I would expect endemic species to be younger than species in the
larger adjacent regions (the Western Indian and Central Pacific regions, respectively).
To address my aims, I constructed the most comprehensive assemblage of coral reef fish
genera with near-complete taxon sampling to-date, and combined it with available fossil data to
simultaneously infer the phylogeny and estimate divergence times of species from four major
coral reef fish families: Chaetodontidae, Labridae, Pomacanthidae and Epinephelidae. I used
this phylogenetic hypothesis in conjunction with recent distributional data to investigate how
the age structure of coral reef fish species varies in response to differing geological histories
among marine biogeographical areas, and test the predictions of the ‘Centre-of’ hypotheses. My
specific questions were:
1) How old are coral reef fish species and do their ages differ among major marine
realms or regions?
2) Do peripheral, isolated locations of high endemism support greater numbers of young
endemics; are peripheral locations a source of ‘new’ species?
6.2 Materials and Methods
Methods of Data selection and handling and Phylogenetic analysis and age estimation for
this data chapter follow those detailed in Chapter 2: General Methods, section 2.2 Multi-
Family Phylogeny.
Biogeographical analysis
The recent work of Kulbicki et al. (2013) was used to classify species’ distributions by
realm: Indo-Pacific, and Atlantic; by region: Western Indian (WI), Central Indo-Pacific (CIP),
Central Pacific (CP), Eastern Tropical Pacific (ETP), Western Atlantic (WA), and Eastern
Atlantic (EA); and to identify those species with distributions restricted to the Hawaiian Islands
80
and the Red Sea. Kulbicki et al. (2013) quantitatively delineated biogeographical patterns based
on species composition. They described the ETP as both a realm and region; however, I
considered it only as a region outside of the Indo-Pacific and Atlantic realms.
Age comparison
Phylogenetic reconstruction is based on a bifurcating process that produces an exponential
distribution of age estimates, I thus log-transformed the data to normalise the distribution. I
used two approaches for defining species age with the aim of comparing ages between realms
and regions. First, I considered the estimated divergence time for all nodes in the phylogeny
subtended by extant species as reflective of species age, excluding fossil calibration and
outgroup species. I refer to this as the full-phylogeny approach where the advantage was
achieving a large sample size for comparison among biogeographical areas. There were several
limitations to this approach including the co-variance of ages shared among sister-species,
which may have a levelling effect on any potential differences in age structure if sister-species
are found in adjacent, or different areas.
Another limitation of the full-phylogeny approach was the impact of missing taxa on the
estimation of species’ ages. Phylogenetic methods are only able to estimate the age of the most
recent common ancestor between two species if both species are present in the analysis.
Therefore, taxa missing from the phylogeny, whether extinct or extant, will cause the ages of
sister taxa to be over-estimated. Genera were selected with the aim of minimising the amount of
missing extant species; however, the proportion of missing extant species in the analysis varied
by genus (Table S6.1, Appendix E). Extinct species will have a greater impact on the age
estimation of older lineages if the rate of extinction through time has remained relatively
constant (Rabosky & Lovette, 2008).
To minimize the effects of missing taxa on species age estimation my second approach
considered only sister-species pairs, excluding monotypic genera. I refer to this as the sister-
81
species approach where I assigned estimated divergence times to the species with the smallest
geographical range and did not include its sister in the analysis. This approach eliminated the
levelling effects of co-varying sister-species’ ages and minimized the likelihood of
underestimating the ages of widespread species that may have persisted through successive
peripheral divergence events (i.e. peripheral budding; Chapters 3, 5).
For both approaches, I used bootstrap re-sampling to test whether the mean age of species
from each biogeographical area differed from a random distribution of mean ages. Bootstrap re-
sampling was used because the data did not meet assumptions of independence required by
parametric analyses. Furthermore, the bootstrap analyses could be applied to both age-
estimation approaches. For each realm, region and province I sampled the corresponding
number of species (for all biogeographical areas with more than two constituent species)
randomly without replacement from the full phylogeny and from the full pool of resultant
species following the sister-species approach. I also compared the mean ages of Red Sea and
Hawaiian Island endemic species to corresponding random samples of species restricted to the
WI and CP regions, respectively, to resolve whether endemic species are young relative to non-
endemic species from the same region and thus potentially mark locations of species
origination. Random sampling was repeated 1000 times for each case. I performed a two-tailed
significance test where the observed mean age was considered significantly different from
random if it was outside of the central 95% of the resampled distribution. I repeated the same
randomisations using the mean upper and lower 95% HPD intervals of species age.
Peripheral speciation
To further explore the frequency of peripheral speciation events I fitted a model to the
distribution of untransformed species’ ages across the entire phylogeny and compared it to
models fitted to distributions of untransformed species’ ages for both Red Sea and Hawaiian
Island endemics. All models were fitted to density distributions of 2 Myr intervals. Finally, to
82
gain insight into potential processes underlying endemism and establish patterns of
geographical range distribution among peripherally isolated species, I determined whether
endemic species are contemporarily allopatric or sympatric with regard to their respective
sister-species (Barraclough & Vogler, 2000) or clade (using independent contrasts sensu
Felsenstein, 1985). All statistical analyses were performed using R (R Development Core Team,
2011).
6.3 Results
Phylogenetic analysis
See Chapter 5, section 5.2 Phylogenetic Methods of Age Estimation for results of the
multi-family phylogenetic analysis (see also Table S5.1 and Fig. S5.1, Appendix D).
Age comparisons
Species with distributions confined to the Indo-Pacific and Atlantic realms had strikingly
similar mean ages and variances (using both the full-phylogeny and sister-species approach;
Table 6.1 and Fig. 6.1). Mean species age was expectedly younger under the sister-species
approach (2.9 Myr versus 5.4 Myr for species sampled using the full-phylogeny approach), but
with comparably little variability between realms. No significant differences in the mean ages
of species from the Indo-Pacific and Atlantic realms were detected when compared to
distributions of means from 1000 resampled permutations, for both the full-phylogeny and
sister-species approach.
Little variation was detected in mean species age among regions under both approaches
(Table 6.1 and Fig. 6.1). No significant differences in mean species age compared to the
randomly resampled distributions for all six regions using the full-phylogeny approach, and for
all five bootstrapped regions using the sister-species approach were found (note that
83
bootstrapping was not performed for the CIP region using the sister-species approach because it
did not have more than two constituent species). Overall, both approaches produced similar
patterns (Table 6.1 and Fig. 6.1). The most notable distinction between the two approaches was
observed in the mean age of widespread species in the Indo-Pacific realm. Under the full-
phylogeny approach, the mean age did not differ significantly from the bootstrapped
distribution, whereas the mean age was significantly older than the central 95% of the
resampled distribution using the sister-species approach. A shift in the age structure of species
in the Atlantic realm was detected along with a slight increase in regional variation of mean
species age using the sister-species approach. In general, the sister-species approach produced
ages 0.6–2.7 Myr younger than the full-phylogeny approach, except for the WA region where
the mean age was 4.1 Myr younger using the sister-species approach.
Species endemic to the Red Sea and Hawaiian Islands had similar mean ages to those
restricted to the WI and CP regions, respectively (Table 6.1). The two oldest species in the WI
region are endemic to the Red Sea (oldest species: Larabicus quadrilineatus; median age: 14.7
Ma; 95% HPD: 7.4–23.8 Ma). However, Hawaiian endemic species did not include the oldest
species in the region. No significant difference in the mean age of species endemic to the Red
Sea or Hawaiian Islands was found when compared to the bootstrapped distributions of species
across the full phylogeny, or compared to the bootstrapped distributions of species restricted to
the corresponding regions. These results persisted under the sister-species approach.
Peripheral speciation
Phylogeny-wide age estimates were randomly distributed under a negative binomial model
(r = 0.2, p = 3.2; n = 293; d.f. = 1, 292; χ2 = 2.72; P = 0.10) (Fig. 6.2). However, Red Sea
endemic species have diverged steadily through time and fit a Poisson model (λ = 1; n = 16; d.f.
= 1, 15; X2 = 0.09; P = 0.76) (Fig. 6.3a), while Hawaiian endemic divergences fit neither a
84
Table 6.1 Sample sizes and resulting mean species age for two approaches used to compare the ages of
coral reef fish species between marine biogeographical realms and regions. (a) The full-phylogeny
approach that considered the ages of all extant species, excluding fossil calibration and outgroup species.
(b) The sister-species approach that considered the age of only the species with the smallest geographical
range (per sister-species pair, excluding monotypic genera). Contemporary spatial data were used to
classify species by realm and region, and to identify those species endemic to the Red Sea and Hawaiian
Islands (IUCN, 2011) following the biogeographical delineations of Kulbicki et al. (2013). Regions
included the: Western Indian (WI), Central-Indo Pacific (CIP), Central Pacific (CP), Eastern Tropical
Pacific (ETP), Western Atlantic (WA), and Eastern Atlantic (EA). Numbers in parentheses corresponding
to the WI and CP indicate the number of species from that region endemic to the Red Sea and Hawaiian
Island provinces, respectively. Region-restricted species combined with widespread species (WS) in the
Indo-Pacific and Atlantic realms to give the total number of species for each realm. Bootstrap re-
sampling was performed for all biogeographical areas with more than two constituent species. Asterisks
indicate a significant difference (P < 0.05) in the mean age of species compared to 1000 resampled
permutations.
Mean (Ma) Mean (Ma)
Realm No. Extant Species Age 95% HPD Region No. Extant
Species Age 95% HPD
(a) Full-phylogeny approach
WI 29 (16)
3.9 (4.2)
2.0–6.2 (2.2–6.8)
CIP 6 2.9 1.4–4.9
CP 18 (9)
5.3 (4.8)
2.9–8.6 (2.7–7.4)
Indo-Pacific 204 5.5 3.1–8.4
WS 151 5.9 3.4–9.0 — — — — ETP 22 4.6 2.7–6.9
WA 34 6.3 3.8–9.2 EA 21 5.0 2.6–7.8 Atlantic 57 5.6 3.3–8.5 WS 2 1.9 0.6–3.9
(b) Sister-species approach
WI 15 (9)
2.6 (3.0)
1.1–4.7 (1.2–5.3)
CIP 2 0.5 0.1–1.2
CP 9 (4)
2.6 (1.2)
0.9–4.9 (0.4–2.2)
Indo-Pacific 66 3.0 1.3–5.3
WS 40 3.3* 1.4*–5.9* — — — — ETP 8 2.1 0.9–3.8
WA 10 2.2 0.9–4.0 EA 6 4.4 2.1–7.0 Atlantic 17 2.9 1.3–5.0 WS 1 1.2 0.3–2.4
85
Figure 6.1 Biogeographical variation in the ages of coral reef fish species. (a) Map showing two marine
realms, the Indo-Pacific and Atlantic, and the regions within each realm differentiated by colour. Western
Indian (WI) (red), Central Indo-Pacific (CIP) (orange), Central Pacific (CP) (light orange), Eastern
Tropical Pacific (ETP) (green), Western Atlantic (WA) (light blue), and Eastern Atlantic (EA) (dark
blue). Areas follow those outlined by Kulbicki et al. (2013). Two peripherally isolated provinces within
the Indo-Pacific realm, the Hawaiian Islands and the Red Sea, are outlined in black. The map uses a
Behrmann projection. Below the map are the results of age comparisons of coral reef fish species among
marine realms and regions using (b) a full-phylogeny approach, and (c) a sister-species approach. Mean
species age (dark grey horizontal bars) and 95% highest posterior density (HPD) intervals (light grey
shading) are displayed for species in the Indo-Pacific and Atlantic realms as well as for species restricted
to regions (corresponding coloured circles). Black circles indicate mean ages (and corresponding 95%
HPD intervals) of non region-restricted species for both the Indo-Pacific and Atlantic realms.
Corresponding coloured squares represent median species age. Asterisks indicate a significant difference
(P < 0.05) in the mean age of species compared to 1000 resampled permutations. Sample sizes for each
biogeographical area are listed in Table 6.1.
86
negative binomial, nor a Poisson model. Instead, the age estimates of Hawaiian endemic species
fit a bimodal distribution with distinct peaks between 0–3 Ma and 8–12 Ma (Fig. 6.3b), which
conforms to a mixture of two normal distributions [D = 20.58; D > 2 required for clear
separation; (Ashman et al., 1994)]. Furthermore, the mean 95% HPD intervals of both peaks do
not overlap (95% HPD early peak: 5.7–13.9 Ma; late peak: 0.4–2.3 Ma) (Fig. 6.4). The
separation of Hawaiian Island endemic species’ divergences is apparent despite ongoing
divergence within the CP throughout this period. It also holds true for multiple species within
genera. For example, two of the three endemic Chaetodon species arose in the late peak
(between 0–3 Ma), while the third endemic Chaetodon species arose in the early peak (between
8–12 Ma) (Fig. 6.4).
Figure 6.2 Frequency distribution of age estimates for coral reef fish species across the entire mulit-
family phylogeny plotted as the proportion of total species (left side) and as a frequency (right side) on
the y-axis. This age distribution fits a negative binomial model (r = 0.2, p = 3.2; n = 293; d.f. = 1, 292; χ2
= 2.72; P = 0.10), represented by the black line.
574941332591 17
Pro
porti
on
Estimated species age (Myr)
MiocenePlioPlt Oligocene Eocene Palaeo0.0
0.1
0.2
0.3
0.4
0.5
Freq
uenc
y
0
23
46
69
92
115
138
87
Figure 6.3 Frequency distributions of age estimates for coral reef fish species plotted as the proportion of
total species (left side) and as a frequency (right side) on the y-axis. (a) Species endemic to the Red Sea
(grey bars) fit a Poisson model (λ = 1; n = 16; d.f. = 1, 15; X2 = 0.09; P = 0.76), represented by the black
line; asterisks denote the age distribution of species restricted to the Western Indian region. (b) Species
endemic to the Hawaiian Islands (grey bars) have a bimodal age distribution (Ashman’s D = 20.58);
asterisks denote the age distribution of species restricted to the Central Pacific region.
15131197531Estimated species age (Myr)
Pro
porti
onP
ropo
rtion
MiocenePlioPlt
15131197531Estimated species age (Myr)
MiocenePlioPlt
(a) Red Sea and WI
(b) Hawaiian Islands and CP
0.0
0.1
0.2
0.3
0.4
0.5
0.0
0.1
0.2
0.3
0.4
0.5
Freq
uenc
y
0
2
4
6
8
Freq
uenc
y
0
1
2
3
4
*
*
* *
*
*
*
* * *
88
To explore peripheral speciation processes I determined whether species endemic to the
Red Sea and Hawaiian Islands are contemporarily allopatric or sympatric relative to their sister
taxon. Endemics in these two regions differed in the proportion of allopatrically and
sympatrically distributed sister-species (Fig. 6.4). Seven of the nine Hawaiian endemic species
(78%) have allopatric distributions with their sister-species or clade, while the remaining two
species have sympatric distributions (22%). Six of the nine endemic species are sister to a single
species and the remaining three are sister to a clade. Interestingly, both sympatric species are
sister to a clade and were among the early wave of speciation in the Hawaiian Islands (between
8–12 Ma) with an average degree of overlap less than 10%. The average degree of sympatry for
Hawaiian endemic species was 1%, compared to an average of 19% sympatry among species
restricted to the CP region (where four of nine species were sympatric). In contrast, half of the
Red Sea endemic species (eight of 16 species) have allopatric distributions, and half have
sympatric distributions. Of the 16 endemic species, 11 are sister to a single species, and the
remaining five are sister to a clade. Species that are sister to a clade are distributed throughout
the age range of endemic Red Sea species, as are allopatric and sympatric species. The average
degree of sympatry for Red Sea endemics was 23%, similar to the average degree of sympatry
for species in the WI region (19%, eight of 13 species were sympatric), but substantially larger
compared to Hawaiian endemics.
6.4 Discussion
Despite differing geological and evolutionary histories between the Indo-Pacific and
Atlantic biogeographical realms, the results show limited variation in the ages of their
constituent coral reef fish faunas. The CIP and its peripheral regions have all experienced recent
divergence events, with no detectable difference in the mean ages of reef fish species among
regions in the Indo-Pacific. Atlantic regions have also experienced recent divergence events
with mean ages similar to those of the Indo-Pacific. Furthermore, ages of endemic species do
not differ from more widespread species. Therefore, the results do not support the
89
indiscriminate use of endemic species as markers of species origination. Interesting patterns of
diversification in isolated locations provide insights into the processes underlying endemism
and peripheral speciation. Specifically, I report notable differences between the Red Sea and
Hawaiian Islands in the timing of divergence events and in patterns of contemporary
geographical distribution of endemic species.
Figure 6.4 Ages of Red Sea and Hawaiian endemic coral reef fish species [circles and squares; 95%
highest posterior density (HPD) intervals indicated by horizontal black lines]. Circles represent endemic
species sister to a single species; squares represent endemic species sister to a clade. White shapes
indicate allopatric endemic species; black shapes indicate sympatric endemic species (i.e. degree of
sympatry above zero). Degree of sympatry was calculated following Barraclough & Vogler (2000) and
using independent contrasts (Felsenstein, 1985) for endemic species sister to a clade. The ages of
Hawaiian endemics conforms to a mixture of two normal distributions according to Ashman’s D statistic
[D = 20.58; D > 2 required for clear separation; (Ashman et al., 1994)]. Dashed vertical lines show the
mean of the two separate periods of divergence and the grey shaded area indicates the mean 95% HPD
intervals, which do not overlap (95% HPD early peak: 5.7–13.9 Ma; late peak: 0.4–2.3 Ma).
Haw
aiia
n Is
land
s
Chaetodon
Thalassoma
Chlorurus
Larabicus
Red
Sea
MiocenePlioPlt
20.015.010.05.00.0
Scarus
Chaetodon
Thalassoma
Chlorurus
Scarus
Anampses
90
Age comparisons
The Indo-Pacific and Atlantic realms are undoubtedly characterised by unique assemblages
of coral reef fishes (Briggs & Bowen, 2012; Kulbicki et al., 2013). While some species have
attained circum-tropical distributions, most species are restricted to a particular realm and share
the influence of its historical periods of isolation (Kulbicki et al., 2013). Vicariance events such
as the closure of the Tethys seaway (Steininger & Rögl, 1984) and the rise of the Isthmus of
Panama (Coates & Obando, 1996) have produced diffuse signals of vicariance among reef fish
lineages, while soft hydrological barriers to dispersal can result in tightly concordant vicariant
speciation (Cowman & Bellwood, 2013b). Furthermore, each region has a markedly different
amount of space for reef fishes to occupy, which has varied through time, and may have
influenced the rate of species divergence or extinction (Bellwood & Wainwright, 2002; Renema
et al., 2008). Despite the potential for temporally diffuse or concentrated vicariance events, and
variable habitable area within each realm, few significant differences among the ages of
constituent species were found using two approaches for age comparison. On average, recent
divergence of extant species has occurred throughout the past 1–5 Myr in the Indo-Pacific and
Atlantic realms, suggesting similar timing in the divergence of reef fishes, despite different
geological histories and different contemporary patterns of biodiversity.
Differences between the two approaches in the ages of widespread species in the Indo-
Pacific realm suggest that extinction may have resulted in substantial overestimation of species’
ages using the full-phylogeny approach. While the larger magnitude of change in mean species
age in the WA region, relative to other regions, suggests that extinction may have had the
greatest influence on the ages of extant species in this region. The Atlantic has experienced
substantial loss of marine taxa during a period of faunal turnover in the Plio-Pleistocene
(Bellwood, 1997; O'Dea et al., 2007), which may explain the differences observed when
comparing full-phylogeny and sister-species approaches.
91
In the Atlantic, widespread species are young compared to their region-restricted
counterparts. However, I found just two species with distributions spanning the Eastern and
Western Atlantic regions. The low number of widespread species in the Atlantic may be due in
part to the relatively low overall coral reef fish biodiversity of this realm (Bellwood, 1997;
Kulbicki et al., 2013), but it may also indicate that regional spatial structure is an important
characteristic of Atlantic species (Bender et al., 2013). In comparison to the Indo-Pacific, the
Atlantic has fewer centrally located islands available to facilitate range expansion, which may
produce more defined spatial structure among its faunal constituents.
More variation in the ages of reef fishes was detected between regions than among realms,
with younger ages in the CIP compared to the WI and CP. Younger ages of reef fish species in
the CIP, although not significantly different from random permutations, lends some support to
the Centre of Origin or Centre of Survival hypotheses. Recent models of coral reef fish
evolution and dispersal over the last 65 Myr demonstrate that the IAA (located within the
greater CIP region) has played a number of different roles, supporting the accumulation,
survival, origination and export of species (Barber & Bellwood, 2005; Barber, 2009; Halas &
Winterbottom, 2009; Bowen et al., 2013; Cowman & Bellwood, 2013a). More specifically,
models suggest that since the Miocene (23 Ma) the IAA has been characterised by exceptionally
high rates of species origination (Cowman & Bellwood, 2013a). This may explain the slightly
younger species’ ages obtained for the CIP relative to adjacent regions that, according to the
models, were colonized by lineages that originated in the IAA during this time (Cowman &
Bellwood, 2013a).
If vicariance has played a major role in recent species diversification, it would have a
neutralising effect on any potential patterns between vicariant regions when comparing species’
ages using the full-phylogeny approach. For example, the rise of the Isthmus of Panama
produced geminate species pairs with a shared age that are now distributed either side of the
barrier. Therefore, when comparing the ages of species from the WA and the ETP, the ages of
geminate species pairs are considered for both regions and neutralize any differences in age
92
structure between them. However, patterns of relative age between realms and regions remained
constant, with the exception of species in the WA region, when such co-variance of ages was
accounted for using the sister-species approach.
The methods employed here considered only extant species and their lineages, and therefore
were not capable of resolving historical evolutionary differences that may have distinguished
biogeographical areas. The applied protocols to maximize sampling of extant species meant that
generic sampling achieved within families was limited and I was not able to resolve deeper
splits in the phylogeny with confidence, precluding an assessment of historical differences
among lineages. Instead, I have focused on recent speciation events because it is unlikely that
many older species have survived to the present day. If older species have survived and also
given rise to other species through time (i.e. peripheral budding; Chapters 3, 5), their true age
may be masked by recent speciation events. Thus, historical signals are largely overwhelmed by
recent speciation. Different geological and evolutionary histories among regions have probably
shaped lineages, but speciation has largely been shaped by events in the past 1–5 Myr.
Peripheral speciation
Endemic species in the Red Sea and Hawaii displayed different age distributions when
compared to each other and to the underlying age distribution of the full phylogeny, suggesting
that distinctive processes of diversification have operated at these peripherally isolated
locations. The topological and chronological hypotheses of Red Sea and Hawaiian endemic
species agree with previous phylogenetic hypotheses for Chaetodon (Fessler & Westneat, 2007;
Bellwood et al., 2010; Craig et al., 2010), Anampses (Chapter 3), Chlorurus (Choat et al.,
2012), Larabicus (Westneat & Alfaro, 2005; Cowman et al., 2009; Kazancıoğlu et al., 2009),
Scarus (Choat et al., 2012), and Thalassoma (Bernardi et al., 2004), offering additional
confidence in the chronogram.
93
Red Sea endemics appear to have arisen steadily throughout the past 16 Myr, roughly 10
Myr after the sea first appeared (Bosworth et al., 2005). They include the oldest extant lineage
in the WI region, Larabicus quadrilineatus, which is estimated to have diverged 14.7 Ma,
around the time when the Red Sea was becoming increasingly isolated from the Mediterranean
(Bosworth et al., 2005) and the Arabian hotspot was dwindling (Renema et al., 2008). In the
recent geological past, the Red Sea has experienced volatile changes in temperature and salinity
(Biton et al., 2008). The effects of these environmental fluctuations have reportedly caused
mass extirpation of marine organisms including planktonic foraminifera (Hemleben et al.,
1996). The results, as well as other phylogeographic studies of reef fishes (DiBattista et al.,
2013), suggest that the Red Sea, or the adjacent Gulf of Aden, has sustained coral reef fish
lineages (and presumably coral reefs) throughout these environmental fluctuations. Older
endemic lineages, like L. quadrilineatus, are likely to have survived outside of the Red Sea
during extreme environmental periods and subsequently re-invaded when the Red Sea opened
up to the WI region c. 5 Ma and the environment became suitable for the maintenance of coral
reefs (Siddall et al., 2003; Bosworth et al., 2005). The majority of Red Sea endemics (75% of
those studied herein) diverged after this time. The continuity of the continental shelf to the east
and south of the Red Sea, in combination with its relative close proximity to the IAA, has likely
facilitated the ongoing divergence of lineages throughout the past 16 Myr.
In contrast to the Red Sea, the Hawaiian Archipelago is located in the central Pacific Ocean
and is part of the larger Hawaiian-Emperor seamount chain, a series of volcanic islands and
atolls separated by oceanic channels. The data suggest that colonization of the Hawaiian Islands
has occurred independently for multiple species belonging to the genera Chaetodon, Anampses,
and Thalassoma. This is consistent with previous phylogenetic hypotheses (Bernardi et al.,
2004; Fessler & Westneat, 2007; Craig et al., 2010; Chapter 3). Multiple models of
diversification have likely led to the divergence of Hawaiian endemics including, but not
limited to, successive colonization/division and peripheral budding (Chapter 3). For example,
Craig et al. (2010) proposed that a closely related Chaetodon species complex containing
94
Chaetodon punctofasciatus, Chaetodon pelewensis, and Chaetodon multicinctus, originated in
the western Pacific, spread through the South Pacific Islands and finally colonized the Hawaiian
Islands. With the inclusion of the closely related WI species, Chaetodon guttatissimus, I found
that the diversification of species in this clade could fit a successive division or colonization
model, with initial separation between the Indian Ocean and the CIP/CP, followed by separation
between the CIP/CP and the Hawaiian Islands. However, the region of origination of this
lineage remains unclear.
In contrast to the ongoing cladogenesis in the Red Sea, I found evidence for two distinct
waves of divergence among Hawaiian endemics. The two waves of divergence (0–3 Ma and 8–
12 Ma) occurred either side of a broad period of increased primary productivity (Dickens &
Owen, 1999) that coincided with increased cladogenesis for a wide range of taxa in the Indo-
Pacific, including reef fishes, between 3.5 and 9.0 Ma (Renema et al., 2008; Reid et al., 2010;
Cowman & Bellwood, 2013a, b). The first wave coincides with the late Miocene-Pliocene (9–
12 Ma) when deep-water circulation reorganisation occurred as a result of the reduction in
deep-water exchange between the Atlantic and Pacific Oceans through the Panamanian gateway
prior to the emergence of the isthmus (Lyle et al., 1995). These changes in deep-water
circulation caused disruption to large-scale ocean circulation patterns (Butzin et al., 2011).
Further evidence suggests that atmospheric and oceanic circulation intensified about 10 Ma
(Rea & Bloomstine, 1986) and during the glacial periods of the past 1.2 Myr (Hall et al., 2001).
These changes in ocean circulation and intensity may have produced more favourable
conditions that allowed founder populations of reef fishes to reach the Hawaiian Islands and
establish themselves, ultimately leading to the divergence of peripatric populations and the
formation of endemic species.
Different proportions of allopatric and sympatric sister-species among Red Sea and
Hawaiian endemics provides additional support for the operation of distinctive processes of
diversification at these peripherally isolated locations. Red Sea endemics have equal
proportions of allopatrically and sympatrically distributed sister-species, and approximately
95
one-third appeared as sister to a clade. Secondary endemism, where a primary endemic gives
rise to one or more subsequent endemics (Rotondo et al., 1981), appears likely to have operated
in the Red Sea province given the sympatric distribution of the well-supported sister-species
pair: Scarus persicus and Scarus ferrugineus, both Red Sea/Persian Gulf endemics (Choat et al.,
2012). The results, in combination with knowledge of the dynamic geological and
environmental past of the Red Sea, suggests that a number of speciation modes may have
operated through time and that both allopatric and sympatric speciation have likely played a
role in generating Red Sea endemics.
While a number of speciation modes may have also led to the diversification of Hawaiian
endemics, the results revealed potential key differences between the two isolated provinces. As
in the Red Sea, one-third of Hawaiian endemic species are sister to a clade. However, allopatric
distributions constitute the bulk (78%) of endemic species’ distributions in the Hawaiian
Islands. With such a low level of sympatry it seems unlikely that either sympatric speciation or
secondary endemism have been important in the evolution of Hawaiian endemic reef fishes.
Rather, allopatric speciation, likely in the form of peripatric speciation, appears to have been the
dominant mode in generating Hawaiian reef fish endemics. The topological relationships and
estimated ages of sympatric endemic species (between 8–12 Ma), combined with the low levels
of average overlap (< 10%), suggest that peripheral budding (sensu Chapters 3, 5) may be a key
model under which Hawaiian endemic reef fishes have diverged.
6.5 Conclusions
Both marine realms and all six biogeographical regions show similar patterns of recent
species divergence. Two peripherally isolated locations with high levels of endemism, the
Hawaiian Islands and the Red Sea, show contrasting patterns. The age structure of Red Sea
endemics is comparable with the larger WI region, with a steady increase in species over the
past 16 Myr. In contrast, the Hawaiian Islands showed two distinct periods of divergence,
96
between 0–3 Ma and 8–12 Ma. Spatial distributions of endemic species relative to their closest
sister taxon reveal that allopatric and sympatric speciation are likely important processes in the
divergence of endemic species, with peripatry especially important in the divergence of
Hawaiian Island endemics. Taken together, the differences in age structure and spatial patterns
between Red Sea and Hawaiian Island endemic species point to markedly different processes of
diversification at two peripherally isolated locations.
97
Chapter 7: The geography of speciation in coral reef fishes Submitted for publication (2014)
7.1 Introduction
Present-day patterns of biodiversity result from evolutionary processes of genetic
divergence, speciation, range-size transformation and extinction. The influence of geography on
these processes has intrigued biologists and biogeographers for centuries (Darwin, 1859; Mayr,
1954). Geographical areas of concentrated diversity, or biodiversity hotspots, provide ideal
opportunities to study evolutionary history and examine the role of geography in speciation. In
the marine realm, a prominent biodiversity hotspot, often referred to as the Coral Triangle, is
found in the central Indo-Pacific (Rosen, 1984; Hoeksema, 2007). The area is a hotspot of
diversity for many marine organisms including plants, invertebrates, and coastal reef-associated
fishes (reviewed in Hoeksema, 2007; Renema et al., 2008; Bellwood et al., 2012). Of the
organisms aligned with this hotspot, the reef fishes are exceptionally diverse with well-
described, interesting distributional patterns.
Many researchers have evaluated the influence of geography on the divergence of ancestral
reef fish lineages (Renema et al., 2008; Cowman & Bellwood, 2013a, b) and the foundations of
reef fish biodiversity are generally well established (Bellwood et al., 2012). Studies of
population connectivity have also illuminated the role of geography in producing and
preserving genetic diversity within species (reviewed in Carpenter et al., 2011; Bowen et al.,
2013). Yet, comparatively little is known about the role of geography in generating and
maintaining diversity at the species level (but see Rocha & Bowen, 2008; Gaither & Rocha,
2013). Because the diversity of life is most often described using species as the focal unit, it is
critically important to establish key factors that influence the diversity of species and the
98
processes by which they diverge. Our ability to predict patterns that may result from global
environmental change crucially hinges on this knowledge.
Uncertainty surrounding sister-species relationships and their associated timing of
divergence is a fundamental reason why the geography of speciation is not yet fully understood.
Without these two key pieces of information, it is difficult to eliminate the importance of past
geological and geographical events in generating and maintaining new species. Traditionally,
sister-species have been defined based on morphological characteristics, including differences
in colour patterns and morphometrics. However, inter-specific relationships are not always as
clear as morphology or colour patterns suggest (Gaither et al., 2014). Molecular phylogenetics
has proved to be a useful tool in establishing sister relationships among taxa; although,
phylogenetic methods require complete sampling of all or most extant, nominal taxa to establish
robust sister relationships (Barraclough & Nee, 2001). Few phylogenies exist for reef fishes
with such complete sampling at the species level (but see Klanten et al., 2004; Bernardi et al.,
2008; Rocha et al., 2008; Crow et al., 2010; Chapters 3, 4 for examples).
Fortunately, molecular data are becoming increasingly available and it is now possible to
evaluate the geography of speciation in reef fishes using a newly constructed phylogenetic
hypothesis that incorporates genera with near-complete sampling of extant, nominal species
across four major reef fish families: Chaetodontidae, Labridae, Pomacanthidae and
Epinephelidae (methods detailed in Chapter 2). The aim of this chapter is to combine robust
sister-species relationships from these phylogenies with precise data detailing species
geographical distributions to investigate the geography of recent speciation events and identify
areas that are important in the generation and maintenance of coral reef fish diversity.
The Central Indo-Pacific (CIP) is a broad area (approximately 90°–165° E) that was
recently quantified as a biogeographical region based on faunal similarity among reef fishes
(Kulbicki et al., 2013). The CIP contains a long-recognised biodiversity hotspot that has been
ascribed various names and delineations dependent upon diversity patterns of different
99
organismal groups and different methods of assessing diversity (Bellwood et al., 2012; Briggs
& Bowen, 2013). Most of the delineations generally describe the area bounded by the
Philippines, the Malaysian Peninsula and New Guinea, with designations including the Coral
Triangle (Allen, 2008; Veron et al., 2009), among others (reviewed in Hoeksema, 2007). The
Indo-Australian Archipelago (IAA) (Bellwood & Hughes, 2001; Renema et al., 2008),
however, encompasses a larger area that is inclusive of the majority of the other delineated
areas, but extends considerably north and south (approximately 90°–160° E, and 10° S–15° N;
Connolly et al., 2003). To consider the largest range of potential processes and factors that
could explain the high level of species richness within the hotspot, this work uses the CIP
region to describe its general location.
Multiple hypotheses have been put forth to explain the bull’s-eye pattern of biodiversity in
the CIP (reviewed in Bellwood et al., 2012). Together, they are often referred to as the ‘Centre
of’ hypotheses. They include the Centre of: (1) Origin, (2) Accumulation, (3) Survival, and (4)
Overlap. Each hypothesis explicitly predicts the evolutionary processes responsible for the
concentration of species richness in the CIP. The Centre of Origin hypothesis, initially proposed
by Ekman (1953) and expanded upon by others (Potts, 1985; Briggs, 1999), asserts that species
originate in the CIP and radiate outward. Species origination is thought to be driven by the
geological complexity of the region, its dynamic history, and the heterogeneous nature of its
environments (McManus, 1985; Hoeksema, 2007; Rocha & Bowen, 2008). The Centre of
Accumulation hypothesis contrasts with the aforementioned by proposing speciation in
locations peripheral to the hotspot and subsequent accumulation of species via inward range
movements (Ladd, 1960) or tectonic collisions (Rosen, 1984). The Centre of Survival
hypothesis places less emphasis on the location of species origination and instead, emphasizes
the CIP as an area of increased species survival (Heck & McCoy, 1978; Barber & Bellwood,
2005). As a consequence, areas outside of the CIP are predicted to have higher rates of
extinction (Potts, 1985; Paulay, 1990; Bellwood & Hughes, 2001). Finally, the Centre of
Overlap hypothesis arose from the recognition of congruent biogeographical boundaries
100
separating closely related taxa (Woodland, 1986; Blum, 1989; Wallace et al., 2000). This
hypothesis argues that distinctive Indian and Pacific Ocean faunas overlap in the CIP, creating
local species enrichment. It invokes changes in sea level and the formation of land barriers as
isolating mechanisms. Importantly, the Centre of Overlap hypothesis predicts distributional
overlap among sister-species within the CIP as a result of secondary contact following periods
of isolation. This differs from the Centre of Accumulation hypothesis, which makes no
predictions about distributional relationships among sister-species.
Testing these hypotheses has proved difficult because in many cases they make identical
biogeographical predictions (Pandolfi, 1992), although they need not be mutually exclusive.
Biogeographical modelling has shown that the CIP has acted sequentially and simultaneously as
a centre of accumulation, survival, and origin of reef fish lineages throughout evolutionary
history (Cowman & Bellwood, 2013a). Equipped with a phylogenetic hypothesis detailing the
evolutionary history of reef fish species, it is now possible to explore factors that have shaped
patterns of total species richness in different biogeographical areas, including the CIP. One
approach is through an evaluation of sister-species and their corresponding biogeographical
distributions.
Sister-species may be either sympatric (i.e. their geographical distributions have some
degree of overlap) or allopatric. If the distributional overlap of sympatric sister-species were
concentrated in the CIP this would support the Centre of Overlap or Centre of Origin
hypotheses. Similarly, if allopatric sister-species have shared vicariant signals associated with
previously described biogeographical barriers in the CIP, this would lend further support to the
suggestion that Pacific and Indian Ocean reef fishes were isolated as a consequence of divisions
within the CIP. The biogeographical analysis of sister-species will therefore enable the
identification of areas important in the generation and maintenance of modern coral reef fish
biodiversity.
Specifically, this study addressed the following questions:
101
1) What is the geographical pattern of overlap among sympatric sister-species? and
2) What is the relative importance of previously described biogeographical barriers in
the divergence of contemporary reef fish species?
7.2 Materials and Methods
Methods of Data selection and handling and Phylogenetic analysis and age estimation for
this data chapter follow those detailed in Chapter 2: General Methods, section 2.2 Multi-
Family Phylogeny.
Biogeographical analysis
Sister-species distributions were classified as either contemporarily allopatric with non-
overlapping distributions (degree of sympatry = 0), or sympatric with some amount of
distributional overlap (degree of sympatry > 0). Degree of sympatry was calculated following
Barraclough and Vogler (2000), where the area of overlap was divided by the smallest range
area between two sister-species. This value extends from zero to one, with increasing values
signifying increasing overlap. Overlapping areas were mapped for all sympatric sister-species
pairs by intersecting their distributional areas and isolating all areas of overlap using GRASS
(GRASS Development Team, 2011) and QGIS (QGIS Development Team, 2012) (Fig. 7.1a). A
heat-map was created using the isolated areas to show locations of concentrated overlap.
Isolated areas of overlap were also grouped according to the quantitatively delineated
biogeographical areas defined by Kulbicki et al. (2013). These included the Indo-Pacific and
Atlantic realms, and the Western Indian (WI), Central Indo-Pacific (CIP), Central Pacific (CP),
Eastern Tropical Pacific (ETP), Western Atlantic (WA), and Eastern Atlantic (EA) regions.
Bootstrap re-sampling was performed in R (R Development Core Team, 2011) to test whether
102
the mean degree of sympatry for species with overlapping distributions in different
biogeographical areas differed from random distributions of mean values. For each realm and
region, the corresponding number of overlapping sister-species pairs was sampled randomly
without replacement from the full set of sympatric sister-species pairs (for all biogeographical
areas with more than two constituent overlapping areas).
The nearest linear distance separating allopatric sister-species’ distributions was also
calculated and mapped. The line was then rotated 90 degrees and mapped as an indication of the
location of vicariance between the allopatric sister-species (Fig. 7.1b). Hereafter these lines are
referred to as splits. Splits were compared to previously described biogeographical barriers
(Bellwood & Wainwright, 2002; Rocha et al., 2007; Floeter et al., 2008; Luiz et al., 2012) to
identify locations important in recent vicariance events. Only those previously described
barriers with three or more corresponding splits were considered for further analysis. All maps
were created using a Behrmann projection.
Distributional symmetry was also calculated for all sister-species pairs (allopatric and
sympatric) as the smallest distribution divided by the sum of the two sister-species’
distributions (Barraclough & Vogler, 2000). This value ranges from zero to 0.5, with increasing
values signifying increasing distributional symmetry. Further bootstrapping was used to test
whether the mean degree of distributional symmetry among allopatric and sympatric sister-
species differed from random distributions of mean values generated from the full pool of
sister-species, and from the respective pools of allopatric and sympatric sister-species.
Symmetry values for sympatric sister-species were grouped by biogeographical region (as in the
previous bootstrap analysis of sympatry), while symmetry values for allopatric sister-species
were classified based on common areas of vicariance. Random sampling was repeated 1000
times for each case. A two-tailed significance test was performed where the observed mean was
considered significantly different from random if it was outside of the central 95% of the
resampled distribution.
103
Figure 7.1 Schematic showing how overlap and vicariance were mapped for sister-species pairs. (a)
Sister-species with a degree of overlap greater than zero were considered sympatric (species S1 and S2).
Degree of overlap was calculated following Barraclough and Vogler (2000), where the area of overlap
(shaded grey) was divided by the smallest range area between the two sister-species. This value extends
from zero to one, with increasing values signifying increasing overlap. Overlapping areas were mapped
for all sympatric sister-species pairs by intersecting their distributional areas and isolating all areas of
overlap. A heatmap was created using these isolated areas to show locations of concentrated overlap. (b)
The nearest linear distance separating allopatric sister-species’ distributions (S3 and S4) was calculated
and mapped as a straight line (black). The line was rotated 90 degrees (grey) and mapped as an indication
of the location of vicariance between the two sister-species. Grey lines were assessed for commonness
among themselves and among known biogeographical barriers to determine locations important in recent
vicariance events.
7.3 Results
Phylogenetic analysis
See Chapter 5, section 5.2 Phylogenetic Methods of Age Estimation for results of the
multi-family phylogenetic analysis (see also Table S5.1 and Fig. S5.1, Appendix D). The
resultant chronogram contained 93 pairs of sister-species, excluding monotypic, fossil
calibration, and outgroup taxa. Thirty-three sister-pairs had allopatric distributions and the
remaining 60 sister-pairs were sympatric.
S3
S1
(a) (b)
S4
S2
104
Vicariance
Multiple splits between allopatric sister-species corresponded to six distinct
biogeographical barriers (Fig. 7.2a, b). These included the Amazon and Orinoco barriers,
Isthmus of Panama (IOP), Hawaiian Archipelago, Indo-Pacific, and a previously unnamed
barrier I term the Mid-Indian Ocean Barrier (MIOB; f in Fig. 7.2a, b). In total, splits at these
barriers accounted for ~61% of all allopatric sister-species. The remaining splits (n = 13) did
not show congruence and often did not align with previously described barriers (Fig. S7.1,
Appendix F). The area of likely vicariance was ambiguous for most of these remaining splits
because the distance between their sister-species distributions was large enough to span
multiple biogeographical barriers. However, some of the remaining splits did solely correspond
to previously described biogeographical barriers, including: II, n, o, and q (see Fig. 7.2a for
location of previously described barriers, remaining splits mapped in Fig. S7.1, Appendix F).
Allopatric sister-species (n = 33) had significantly lower than expected distributional
symmetry compared to the bootstrapped distribution of mean values drawn from all sister-
species pairs. This appeared to be driven largely by sister-species that occur either side of the
MIOB, as these were the only pairs that showed significantly less than expected distributional
symmetry when compared to random sampling from the pool of allopatric sister-species (Fig.
7.2c). Sister-species either side of the Hawaiian Archipelago biogeographical barrier had
similarly low distributional symmetry; however, they were fewer in number and not
significantly different from the bootstrapped distribution of mean values.
Sympatry
The concentration of overlap among sympatric sister-species forms a striking pattern. The
highest concentration of overlap was found in the CIP region in an area concordant with the
biodiversity hotspot (Fig. 7.3a, b). The concentration of overlap declines markedly in all
outward directions in a similar manner to the concentration of biodiversity. The CIP region also
105
contained the largest number of overlapping distributions restricted to a single region (15 of
39), followed by the WI region with seven. However, most distributional overlap was not
restricted to any one biogeographical region, but spanned the Indo-Pacific realm. Sister-species
pairs with realm-wide overlap had significantly greater than expected distributional overlap and
symmetry compared to the full pool of sympatric sister-species (Fig. 7.3c). Sister-species pairs
with overlapping distributions in the Atlantic regions were similar in their comparatively high
degree of sympatry (significantly greater than expected in the WA) and distributional symmetry
(significantly greater than expected in the EA). In contrast, sister-species with overlapping
distributions in the CIP had significantly lower than expected distributional overlap and
symmetry (Fig. 7.3c). The WI region also showed significantly low distributional symmetry and
a similarly low degree of sympatry, although, the latter was not significantly different from the
bootstrapped distribution.
Overall, distributional symmetry among sympatric sister-species (n = 60) was significantly
higher than the bootstrapped distribution of mean values drawn from all sister-species pairs.
Sympatric sister-species with a high degree of sympatry generally also had high distributional
symmetry and vice versa.
7.4 Discussion
The present study identified important areas of common overlap and vicariance for sister-
species pairs of coral reef fishes. This information is consistent with suggestions that
biogeographical boundaries have played an important role in creating and maintaining patterns
of extant species diversity. A high concentration of sympatric distributions restricted to the CIP
region and concordant with hotspot of biodiversity suggests that this area may harbour
sufficiently complex environments to facilitate reproductive isolation that allow closely related
species to co-occur. However, the significantly lower than expected degree of sympatry among
sister-species in this region suggests that it may, more frequently, be an area of secondary
106
107
Figure 7.2 Patterns of vicariance among sister-species of coral reef fishes. (a) Previously described
biogeographical barriers used to assess splits obtained from mapping the nearest linear distance between
allopatric sister-species (figure modified after Bellwood & Wainwright, 2002; Rocha et al., 2007; Floeter
et al., 2008; Luiz et al., 2012). (b) Three or more concordant splits between allopatric sister-species are
mapped as coloured lines in relation to corresponding biogeographical barriers. Colours corresponded to
six distinct biogeographical barriers including the: Mid Indian Ocean (f, red lines), Indo-Pacific (g,
yellow lines), Hawaiian Archipelago (m, green lines), Isthmus of Panama (III, orange line), and the
Amazon and Orinoco (b and a, blue lines) barriers. The remaining allopatric splits (n = 13) are shown in
Figure S7.1, Appendix F. (c) Average degree of distributional symmetry for allopatric sister-species that
occur either side of each biogeographical barrier shown in part (b). The average degree of symmetry for
allopatric splits mapped in Figure S7.1, Appendix F were grouped and plotted as ‘other’. Bootstrapping
was used to test whether the mean degree of distributional symmetry differed from random distributions
of mean values drawn from the full pool of allopatric sister-species. Asterisks denote significant values.
contact between species that primarily occupy adjacent ocean basins. It thus appears to act more
as a Centre of Overlap (and accumulation and survival) than a Centre of Origin for recently
diverged species. Interestingly, allopatric sister-species that were separated by the Indo-Pacific
biogeographical barrier showed similarly low levels of distributional symmetry compared to
sympatric sister-species in the CIP. The WI region and the MIOB were also associated with
highly asymmetrical sister-species pairs, be they allopatric or sympatric, suggesting that these
areas might be important in peripatric speciation. The most prominent vicariant barrier,
however, seems to lie outside of the IAA prompting a re-evaluation of the factors that may
influence the nature and strength of the MIOB in separating reef fish species.
Vicariance
Of the three major biogeographical barriers, and 17 other previously described barriers, the
results of this biogeographical analysis highlight the influence of six barriers on recent reef fish
divergence. These include one of the three major barriers, the IOP, and five soft barriers, the:
MIOB, Indo-Pacific, Hawaiian Archipelago, and the Amazon and Orinoco barriers (Fig 7.2b).
108
109
Figure 7.3 Patterns of sympatry among sister-species of coral reef fishes. (a) Heatmap showing the
concentration of overlap among sympatric sister-species. Biogeographical areas are outlined following
Kulbicki et al. (2013) and included the Atlantic and Indo-Pacific realms; and the Western Indian (WI),
Central Indo-Pacific (CIP), Central Pacific (CP), Eastern Tropical Pacific (ETP), Western Atlantic (WA),
and Eastern Atlantic (EA) regions. (b) Enlarged view of the CIP region showing the highest
concentration of overlapping areas. (c) Average degree of sympatry and distributional symmetry for
sympatric sister-species with overlapping distributions restricted to different regions. Those sister-species
with overlapping distributions not restricted to a particular region were grouped by realm. Bootstrap re-
sampling was used to test whether the mean degree of sympatry and the mean degree of distributional
symmetry differed from random distributions of mean values drawn from the full pool of sympatric
sister-species. Asterisks denote significant values.
Despite the repeated significance of the Terminal Tethyan Event (TTE; ‘I’ in Fig. 7.2a) and the
East Pacific Barrier (EPB; ‘II’ in Fig. 7.2a) in the evolution and biogeography of marine
organisms, including reef fishes (Bellwood et al., 2004; Bernardi et al., 2004; Lessios &
Robertson, 2006; Bellwood et al., 2010; Cowman & Bellwood, 2013b; Chapter 4), these
barriers did not appear to play a major role in the vicariance of recently diverged extant species.
The timing of the TTE [12–18 Ma (Steininger & Rögl, 1984)] places its influence much deeper
in the evolutionary history of reef fishes and away from vicariance of contemporary species
whose ages typically range between 1 and 5 Ma (Chapter 6). Consequently, the influence of the
TTE was in dividing genera and major clades of reef fishes and its lack of influence on the
divergence of present-day species was not surprising.
Unlike the TTE and IOP, the EPB does not have an associated specific time during which it
is expected to have influenced vicariance. Rather than hard, physical landmasses to impede
dispersal, the EPB consists of a large area of deep water (~5000 km) separating the tropical
eastern Pacific from the central Pacific Ocean. It is considered to be the widest marine
biogeographical barrier. Yet, this vast expanse of ocean had little apparent impact on vicariance
of extant species, with only one split solely associated with this barrier (Fig. S7.1, Appendix F).
Because of its soft nature, the EPB is semipermeable and acts as a filter to dispersal and genetic
110
connectivity in a largely unidirectional manner from west to east (Lessios et al., 1998; Lessios
& Robertson, 2006). Thus, it primarily creates geographical range boundaries for some species
and populations, but not in a way that consistently produces isolated populations on either side
that would eventually lead to vicariant species. If populations do become isolated in such a way
that may lead to vicariance, those to the east of the barrier are isolated in a region with
comparatively low species richness (Kulbicki et al., 2013), suggesting that the survival of a
species in this region may be more difficult, or that the dispersion of fauna from the CP has
been sporadic.
The IOP is the most recent hard barrier to have separated populations of marine organisms,
and the timing of its rise [3.1–16 Ma (Coates & Obando, 1996)] falls within the range of extant
reef fish species divergence (Chapter 6). The IOP has been implicated in clade divergence for a
number of reef fish genera (Barber & Bellwood, 2005; Alva-Campbell et al., 2010; Chapter 4).
Thus, the detection of a vicariant signal across this barrier was not unexpected. However,
considering this is a hard barrier to dispersal, the strength of its vicariant signal was surprisingly
weak relative to other soft barriers. This may be the result of subsequent events. Previous
studies of deeper phylogenetic lineages of reef fishes have shown that hard barriers, including
the IOP, can produce temporally diffuse vicariance (Cowman & Bellwood, 2013b). If
subsequent species divergence within the Atlantic or ETP occurred after geminate species pairs
were separated by the land bridge, the present analysis of sister-species pairs would not detect
the IOP as a vicariant barrier. Thus, it appears that although the IOP has influenced extant
species diversity, given the timing and duration of its rise, it likely had a greater impact on the
divergence of clades within reef fish genera than on recent species divisions.
Of the 17 other previously identified biogeographical barriers (Bellwood & Wainwright,
2002; Rocha et al., 2007; Floeter et al., 2008; Luiz et al., 2012), five were concordant with
vicariance between sister-species in the present study. The most prominent barrier detected
herein separates the WI region from the CIP region. It is one of the least discussed barriers in
the literature. The Mid-Indian Ocean Barrier (MIOB) offers few indications of the physical
111
forces underlying vicariance in this location, but it is clear from the results herein, and the work
of others (Winterbottom, 1986; Springer, 1988; Blum, 1989; Springer & Williams, 1994;
Randall, 1998), that it is important in separating sister-species.
The northward movement of the Indian sub-continent has been suggested as the
establishing mechanism underlying the MIOB (Springer, 1988). This movement may have
caused vicariance among ancient lineages of reef fishes, but as noted by Briggs (1989), it was
unlikely to have influenced separation at the species level. Based on the age estimates now
available for contemporary reef fish species (1–5 Ma; Chapter 6), it is conclusive that the
movement of the Indian continent between 50 and 148 Ma cannot explain vicariance among
recently diverged species. What, then, drives the separation of sister-species at this location?
Previous studies have noted that India’s continental shelf areas are largely devoid of coral
reefs (Springer, 1988; Bakus et al., 2000). The scarcity of reefs in this region is generally
attributed to its inhospitable nature as a result of extensive sandy coasts, numerous river deltas
and consequent freshwater plumes that decrease salinity and generate intense siltation
(Springer, 1988; Bakus et al., 2000). It follows that coral reef fishes may also be inhibited by
these same physical factors (Vivekanandan et al., 2012), or are generally scarcer due to the lack
of suitable habitat. Any of these environmental factors may cause a barrier to the dispersal and
genetic connectivity of reef fishes. However, it is proposed herein that the physical
consequences of large sediment loads may be primarily responsible for the strength of this
biogeographical barrier.
The Ganges-Brahmaputra-Megna River delta spans approximately 350 km, discharging into
the Bay of Bengal and contains an unusually high amount of sediment compared to other major
rivers of the world (Curray et al., 2003). A large portion of the sediments are derived from the
fast-eroding Himalaya and have formed an extensive alluvial plain, a large delta system and the
largest submarine fan in the world, the Bengal Fan (Singh, 2008). The Bengal Fan was formed
by turbidity currents where sediment-laden water moves rapidly down a slope because of its
112
higher density relative to the surrounding water. These currents can move exceptionally fast and
velocity estimates for turbidity currents in the Bengal Fan range from 4–10 m/s (Curray et al.,
2003). Current speeds of this nature could easily sweep away any viable marine larvae that
happened to be caught at the surface, in the vicinity of discharge onto the continental shelf.
Thus, if a propagule managed to survive the generally inhospitable nature of the water due to
low salinity and high sediment content, it would likely not survive being swept to depth in a
turbidity current. The influence of turbidity currents as a barrier to dispersal may be exacerbated
if the pelagic larval phase of newly spawned reef fishes coincides with the expansion of the
river plume during warmer monsoonal months.
Sedimentary analysis of the Bengal Fan traced the first deposition of sediments to the early
Eocene; around the time the Indian continent collided with Asia (Curray et al., 2003). The same
study also revealed that the fan is still active and received large volumes of sediment during
periods of lowered sea level in the Pleistocene (Curray et al., 2003), a time when many extant
reef fish species were diverging (Chapter 6) and taxa in the CIP were expanding into the Indian
Ocean (Cowman & Bellwood, 2013a). Most of the splits between sister-species detected in the
present study span the Arabian Sea, Bay of Bengal and the Andaman Sea, encompassing the
entire coastline of India. Therefore, it may be inappropriate to attribute the vast distance
between sister-species solely to turbidity currents on India’s far eastern coast. Rather, there may
be a number of barriers acting independently, or in concert, that contribute to this vicariance.
Interestingly, the Indus River, which flows through Pakistan, India and China and drains
into the Arabian Sea over an area of approximately 210 km, is also one of the most sediment-
laden river deltas on earth. Like the Ganges, the Indus now derives the bulk of its sediment load
from the Himalaya, and has formed an associated submarine fan (Clift & Blusztajn, 2005).
However, approximately five million years ago major rivers of the Punjab rerouted into the
Indus, increasing its drainage and erosional products from the Himalaya (Clift & Blusztajn,
2005). Around the same time (over the past 4 Myr), the Himalaya have experienced faster
erosion as a result of a stronger monsoon (Zhisheng et al., 2001). Turbidity currents in these
113
two sediment-laden river systems, in combination with the increased strength of the monsoon
over the past 4 Myr may have underpinned the strength and magnitude of this biogeographical
barrier.
Symmetry between allopatric sister-species across the MIOB was among the lowest
detected. The barrier between the CP and the Hawaiian Archipelago had similarly low
symmetry among its fewer allopatric sister-pairs. The boundary separating the Hawaiian
Archipelago is likely to be maintained by the relative isolation of the islands compared to the
bulk of reef fish and coral diversity in the CIP (Chapter 6, Randall, 1998; Allen, 2008; Gaither
et al., 2011b). However, the boundary is dynamic and has been breached, with founder events
that can lead to peripatric speciation (Hourigan & Reese, 1987; Jokiel, 1987; Kay & Palumbi,
1987; Chapters 3, 6). Asymmetry in the distributions of species that inhabit the Hawaiian
Islands and the WI region, to the west of the MIOB, suggests that both areas are likely
important locations for peripheral speciation, but with differing temporal patterns of divergence
(Chapter 6). Neither of these areas, however, exclusively receive peripatric species because
genetic connectivity and dispersal also occurs from the Hawaiian Islands toward the CIP (Eble
et al., 2011) and from the isolated Red Sea province toward the larger WI region (DiBattista et
al., 2013).
The second most prominent vicariance signal was detected in the CIP. Barrier ‘g’, as it
appears in Figure 7.2, was first delineated in Blum’s study of closely related chaetodontid
species (1989). Later, Bellwood and Wainwright (2002) included this barrier in their discussion
of the history and biogeography of reef fishes, but without referring to it by a particular name.
Rocha et al. (2007) also identified a barrier in this location and referred to it as the Sunda Shelf
barrier based on the work of Randall (1998). However, a plethora of other barriers have also
been described within the broad CIP region including the Sunda shelf (Fleminger, 1986;
Woodland, 1986; Springer, 1988; Springer & Williams, 1994; Barber et al., 2000; Reid et al.,
2006). The area encompassing these barriers is often referred to generally as Wallacea (Mayr,
1944), and division in the area is attributed to the Indo-Pacific Barrier (Briggs, 1974; Gaither et
114
al., 2010). Barrier ‘g’ stands apart from these barriers in that it lies not within the biodiversity
hotspot (regardless of its designation), but along its western boundary with the Indian Ocean. It
may therefore be regarded as the Sunda Margin Barrier, reflecting the Sunda affinities
highlighted by Rocha et al. (2007), but emphasising its marginal location, a position which
points to different, and potentially older, biogeographical influences than those that supported
vicariance within the broader IAA and Sunda Shelf regions.
Despite the array of described barriers within the CIP, the majority of their associated
vicariance has been attributed to Plio-Pleistocene changes in sea level (reviewed in Hoeksema,
2007; Carpenter et al., 2011; Bellwood et al., 2012). Others have identified the northern
movement of Australia [over the past 120 Ma; (Hall, 1998)] (Woodland, 1986), or regional
tectonic activity (c. 8–16 Ma) that changed surface circulation patterns in the Indian and Pacific
Oceans (Springer & Williams, 1994) as potential causes of vicariance within the CIP. However,
these events are more likely to have influenced vicariance among ancient lineages of reef fishes.
The ages of extant reef fish species (1–5 Ma; Chapter 6) suggest that recent events, such as sea
level changes, may provide a more plausible explanation for extant species-level vicariance.
Barriers between sister-species mainly track the boundary between the Sunda and Sahul
shelves and the Indian Plate. These shelves formed hard barriers during periods of lowered sea
level and potentially restricted all but the most resilient species to their respective oceans
(Bowen et al., 2001; Horne et al., 2008; Gaither et al., 2010; Reece et al., 2010). Recent sea-
level oscillation has also been implicated in intra-specific genetic isolation (Bay et al., 2004;
Klanten et al., 2007; Horne et al., 2008; Leray et al., 2010; Gaither et al., 2011a), which
signifies a degree of vicariance that may lead to geminate species. However, the oscillating
nature of sea level also allows for the overlap of adjacent Indian and Pacific Ocean faunas
(Rocha et al., 2007; Briggs & Bowen, 2013). Evidence for this is discussed below in the
subsection Sympatry. Overall, it seems that the Sunda Margin Barrier (‘g’) has potentially
operated across ecological and evolutionary time scales, with the divergence of extant reef fish
species likely caused by events in the recent past.
115
In the Atlantic, the vicariant signal detected between allopatric sister-species lies between
two previously defined barriers, the Amazon and Orinoco river outflows (b and a, respectively
in Figure 7.2) (Blum, 1989; Hastings & Springer, 1994; Bellwood & Wainwright, 2002). This is
consistent with recent studies, which have shown the freshwater outflows from the Amazon and
Orinoco rivers act as a joint barrier to dispersal spanning 2300 km of the South American
coastline (Rocha, 2003; Floeter et al., 2008; Luiz et al., 2012). The results of the sister-species
analysis expand on the work of others in suggesting that the Amazon-Orinoco barrier has acted
as a filter to dispersal between the north and south-western Atlantic throughout its 11 Myr
history (Hoorn et al., 1995), influencing vicariance between species. This may be particularly
important during periods of low sea level (Kaneps, 1979; Rocha, 2003; Robertson et al., 2006).
However, at times the barrier also allows mixing between populations in the north and south
(Floeter et al., 2008; Rocha et al., 2008). Furthermore, the barrier may impact species
differently depending on their life history characteristics (Rocha et al., 2002).
Overall, vicariance between coral reef fish sister-species has resulted from biogeographical
barriers that have acted at some time over the past 5 Myr. Most of these barriers act as filters to
dispersal, with only one hard barrier having caused detectable vicariance between sister-species.
The permeability and dynamic nature of soft barriers, in combination with founder events have
likely facilitated the majority of vicariance among sister-species leading to the complex patterns
of biodiversity observed today.
Sympatry
For coral reef fishes, the concentrated area of overlap among sympatric sister-species bears
remarkable resemblance to the hotspot of reef fish biodiversity (Allen, 2008), forming a striking
bull’s-eye pattern with the highest concentration of overlap within the CIP and marked decline
in all outward directions. This lends strong support to the Centre of Overlap model that predicts
range expansion of isolated Pacific and Indian Ocean faunas following sea-level rise and
116
subsequent overlap of sister-species within the CIP (Woodland, 1983). However, it may also be
possible for the CIP region to have attained the highest concentration of overlap due to its
position in the middle of three bounded Indo-Pacific regions (Connolly et al., 2003), or because
of the extent of coral reef habitat in the area (Bellwood & Hughes, 2001; Bellwood et al.,
2005), which would facilitate rapid expansion and overlap of ranges. Additionally, if overlap
among sister-species is not largely restricted to the CIP, or if sister-species have widely
overlapping distributions, this may further diminish support for the Centre of Overlap model.
However, the CIP contained the largest number of overlapping distributions restricted to a
single biogeographical region, and these sister-species had significantly less than expected
distributional overlap. This evidence, considered together with corresponding patterns of
vicariance associated with the Sunda Margin Barrier, further highlights the importance of the
CIP as an area of evolutionary dynamism where the separation of Indian and Pacific Ocean
faunas and their subsequent range expansion has led to increased species overlap and higher
regional biodiversity.
The average age of reef fish species, from 1 to 5 Ma (Chapter 6), suggests that tectonic
rearrangements in the Miocene (cf. Renema et al., 2008) need not be invoked as the mechanism
of isolation responsible for separating extant Indian and Pacific Ocean reef fish species in the
CIP (cf. Rocha & Bowen, 2008; Gaither & Rocha, 2013), although geological events of this
nature likely contributed to some vicariance among reef fish lineages. Species diversification is
temporally concordant with the initial suggestion proposed by Woodland (1983) that
Pleistocene low sea level stands may have led to the isolation of populations in the adjacent
ocean basins. The detection of a vicariant signal among allopatric sister-species across the
Sunda Margin Barrier lends further support to the notion that the Sunda and Sahul shelves
influenced species divergence as land barriers during low sea level stands.
Distributional asymmetry between sister-pairs whose overlap was restricted to the CIP
region also supports the Centre of Overlap and Accumulation models. Such distributional
117
patterns may have arisen from the expansion of Pacific or Indian Ocean species into the CIP
region whereby vicariance may have caused peripheral divergence and produced sister-pairs
with asymmetrical distributions. However, if the most geographically restricted species in the
aforementioned example originated in the CIP, then this may also be considered as evidence for
the Centre of Origin model. Sympatric sister-species in the WI region had similarly low levels
of distributional symmetry. Comparatively, the WI region has limited, bounded habitat, which
could account for asymmetry among sister-species whose distributions overlap in the region.
Alternatively, peripatric speciation may be important in producing peripherally isolated species
in this region (Chapters 4, 6). Overall, peripatric speciation appears to play a greater role in
generating allopatric sister-pairs, given the significance of their asymmetric distributions. While
it may also play a role in generating sympatric sister-species in certain regions, subsequent
range expansion likely leads to increasing distributional overlap and symmetry among
sympatric sister-pairs.
The concentration of overlap among sympatric sister-species in the CIP is not inconsistent
with other hypotheses that invoke range contraction of previously widespread species, or range
expansion of species that originated in the CIP region. Over time the dispersal ability of most
reef fishes could produce the patterns detected herein under the latter scenario (Lynch, 1989).
However, multiple lines of congruent evidence from both allopatric and sympatric sister-species
pairs that highlight patterns unique to the CIP region, and the lack of a general explanation for
why such common range contraction may have occurred, would favour the Centre of Overlap
model. Alternate hypotheses are more difficult to evaluate without direct evidence of the
location of species origination. While such records remain elusive, phylogeographic data may
provide insights into processes of genetic isolation preceding species divergence. Indeed,
divergent genetic lineages within many coral reef species complexes also overlap in a manner
consistent with the Centre of Overlap hypothesis (Bay et al., 2004; Craig et al., 2007; Horne et
al., 2008; Gaither et al., 2011a; DiBattista et al., 2012).
118
The majority of sympatric sister-species in the present study had high levels of overlap and
distributional symmetry with ranges often spanning large portions of the Indo-Pacific realm.
This does not necessarily refute evidence from the Centre of Overlap model because high
dispersal potential may be conserved between closely related species enabling them to vastly
expand their geographical ranges during periods of connectivity and increased habitat
availability (Bellwood & Wainwright, 2002). Furthermore, barriers to dispersal do not influence
species in a uniform manner as certain life history characteristics may make species more or
less susceptible to range expansion (Luiz et al., 2012). However, widespread species are
generally older than their regionally restricted counterparts (Chapter 6) and therefore are more
likely to have been impacted by extinction. Thus, widespread, highly sympatric sister-pairs may
be the only extant species remaining in a clade where other, more restricted, species went
extinct or have not yet been described. Consequently, they may not actually be sister-species.
For example, sympatric sister-pairs in the Atlantic show similar patterns of widely symmetrical
overlapping ranges. However, faunal turnover and substantial loss of marine taxa in the Plio-
Pleistocene has been documented in the Atlantic (Bellwood, 1997; O'Dea et al., 2007) and may
have played a key role in establishing the significantly high degree of overlap and distributional
symmetry observed in extant sympatric species in this realm.
Regardless of the mechanisms responsible for the patterns of concentrated overlap among
sister-species within the CIP (Centre of Overlap and other models), evolutionary theory
stipulates that in areas of geographical overlap closely related species must differ in their
ecological niche or mate selection to maintain reproductive isolation, which is required for their
persistence as separate species (Brown & Wilson, 1956; Hardin, 1960). The CIP may
encompass the geological complexity, habitat heterogeneity and reef area required to facilitate
the co-occurrence of closely related genetic lineages (i.e. populations and species) while
maintaining their separation through nice partitioning. Nevertheless, hybridization of reef fish
species in and around the CIP (McMillan et al., 1999; Yaakub et al., 2006; Hobbs et al., 2009),
implies that niche partitioning or mate selection may take some time to develop and establish
119
full reproductive isolation upon secondary contact of closely related species. Alternatively,
these mechanisms may break down at times (Montanari et al., 2012), leading to the production
of hybrid species.
7.5 Conclusions
This work has helped elucidate areas important in the generation and maintenance of
modern coral reef fish biodiversity. Focusing on well-established sister-species and their
relative geographical distributions, I have shown that species vicariance has resulted from
biogeographical barriers that have been in place over the past 5 Myr. The permeability and
dynamic nature of most of these barriers suggest that they act as filters to dispersal in
combination with founder events. Recent expansion of geographical ranges probably
underpinned the distinctive concentration of overlap of sympatric sister-species in the CIP, a
pattern that may be maintained by the geological complexity and extensive heterogeneous
habitat of the region. Together, this evidence largely supports the Centre of Overlap hypothesis
while highlighting the dynamic evolutionary history of reef fish species in the central Indo-
Pacific region.
120
Chapter 8: Concluding Discussion
A framework detailing the chronology of extant species divergence combined with detailed
biogeographical data is essential to investigate evolutionary processes that shape patterns of
biodiversity. This thesis combined molecular phylogenetic approaches with biogeographical
analyses to reconstruct the evolutionary history of extant reef fish species. The development of
methods to minimize the impacts of extinction and successive peripheral speciation on age
estimation have resulted in more precise age estimates for reef fish species (Chapters 5, 6).
These methods are widely applicable to other taxonomic groups and their application may shed
light on previously unrealized processes of divergence for some. Across all biogeographical
areas throughout the world’s tropical oceans, species of coral reef fishes have similar
evolutionary ages (Chapter 6). A combination of evolutionary processes has shaped the
distinctive biogeographical patterns of modern-day reef fishes, including fluctuations in soft
biogeographical barriers (Chapter 7), founder events (Chapters 3–7), and potential divergence
in sympatry (Chapters 4, 6 and 7). As a whole, this thesis has identified common evolutionary
and biogeographical patterns among reef fish species and begun to unravel potential processes
involved in species divergence and maintenance that can be further explored within the
spatiotemporal framework it has provided.
When did reef fish species diverge?
The findings of this research emphasize the need to distinguish between the timing of
species divergence and the evolutionary age of species. Phylogenetic methods considered
within a temporal framework provide estimates of the most recent divergence events between
extant taxa. Applied to two case studies, these methods revealed that species divergence within
121
the reef fish genera Anampses and Pomacanthus occurred predominantly from the mid-Miocene
onward (Chapters 3, 4). Across a broader range of reef fish species from four families
(Chaetodontidae, Labridae, Pomacanthidae and Epinephelidae) temporally-calibrated
phylogenetic hypotheses revealed that both the Atlantic and Indo-Pacific biogeographical
realms experienced divergence events throughout the past nine million years, with two
distinctive periods of divergence among species endemic to the Hawaiian Islands (Chapter 6).
Evolutionary ages of species can be estimated by assembling recently diverged sister-pairs
and applying a biogeographically focused approach to minimize the impacts of historical events
(Chapters 5, 6). The application of this method uncovered a general pattern of geographical
range expansion with increasing evolutionary age among coral reef fish species (Chapter 5).
General evolutionary theories that predict patterns of this nature (Willis, 1922; Wilson, 1961;
Hubbell, 2001; Ricklefs & Bermingham, 2002) have limited supporting evidence from marine
systems and the results presented herein lend them initial support. The evolutionary age of
extant reef fish species ranges from 1 to 5 Ma, with markedly little variation in the ages of
species distributed throughout the world’s biogeographical realms and regions (Chapter 6).
Well-known vicariance events such as the Terminal Tethyan Event and the rise of the Isthmus
of Panama had little to no impact on the divergence of extant reef fish species (Chapter 7),
rather these events produced lineage divisions within genera (Chapters 3, 4).
The chronology of species divergence presented in this thesis contributes to the broad study
of the evolutionary history of reef fishes (reviewed in Bellwood et al., 2012). The evolutionary
ages estimated for species within the Chaetodontidae, Labridae, Pomacanthidae and
Epinephelidae add to the growing work on age estimation of coral reef fish species (McCafferty
et al., 2002; Klanten et al., 2004; Read et al., 2006; Bernardi et al., 2008; Choat et al., 2012;
Gaither et al., 2014). Both temporal components, the timing of species divergence and the
evolutionary ages of species, provide distinct information necessary to eliminate the likely
influence of palaeogeographical events on the creation and maintenance of biodiversity.
122
How are closely related species geographically distributed?
The geographical distribution of closely related species in conjunction with the chronology
of their divergence suggests that successive peripheral speciation, or peripheral budding, may
have generated substantial species diversity among reef fishes (Chapters 3-7). These results add
to the mounting evidence that founder events and peripatric speciation are important in faunal
enrichment within marine systems where few hard barriers to dispersal exist (Paulay & Meyer,
2002; Queiroz, 2005; Floeter et al., 2008; Drew & Barber, 2009; Budd & Pandolfi, 2010;
Gaither et al., 2010; Malay & Paulay, 2010; DiBattista et al., 2011; Fitzpatrick et al., 2011).
Successive peripatric speciation has the potential to generate closely related species with highly
overlapping geographical distributions when the founded area is constrained (Chapters 3, 4, 6,
7). Common vicariance separating sister-species across the Mid-Indian Ocean Barrier (MIOB)
confirmed the marked influence of this previously identified biogeographical barrier
(Winterbottom, 1986; Springer, 1988; Blum, 1989; Springer & Williams, 1994; Randall, 1998;
Bellwood & Wainwright, 2002) on the divergence of contemporary reef fish species (Chapter
7). Like most soft biogeographical barriers (Rocha, 2003; Floeter et al., 2008), the MIOB is
permeable and dynamic. The oscillation of permeable barriers appears particularly important in
facilitating founder events during periods of favourable environmental conditions (Chapters 6,
7).
This thesis highlights the extensive nature of sympatry among closely related reef fish
species (ranging from partial to complete distributional overlap) and emphasizes the necessary
disentanglement of process from pattern by demonstrating that a number of speciation modes
likely led to the divergence of sympatric species (Chapters 4, 7). Specifically, the high
concentration of distributional overlap among sister-pairs concordant with the hotspot of
biodiversity in the central Indo-Pacific supports this region as a Centre of Overlap, where
vicariance followed by geographical range expansion has likely underpinned regional patterns
of overlap and biodiversity (Chapter 7). These findings agree with a large number of studies on
intra-specific genetic divergence of marine organisms (reviewed in Carpenter et al., 2011;
123
Gaither et al., 2011a; DiBattista et al., 2012; Ackiss et al., 2013; DeBoer et al., 2014; Raynal et
al., 2014) and offer support for the notion that the central Indo-Pacific region is an
evolutionarily dynamic area.
What does this suggest about the processes responsible for contemporary patterns of reef fish
diversity?
Overall, this thesis combined a detailed description of the temporal evolution of coral reef
fish species with biogeographical analyses to form a spatiotemporal framework that facilitated
the exploration of evolutionary processes responsible for contemporary patterns of biodiversity.
Rather than the traditional view of a single predominant evolutionary process, this work
supports the notion that multiple, alternate processes have acted in concert to generate and
maintain contemporary patterns of biodiversity. Modern-day reef fish species have temporally
constrained divergence coupled with complex biogeographical patterns. The ability to
investigate the mechanistic processes of evolution at the species level within a spatiotemporal
framework is a necessary first step to the successful prediction of future patterns of biodiversity.
Every effort has been made to reduce potential confounding effects within this thesis;
however, the results must be interpreted with care. The primary concern pertains to taxon
sampling and the probable existence of cryptic, yet to be described species. Future studies may
test the generality of the results by increasing taxon sampling and including any newly
described, or probable biological species (Chapters 5–7). The temporal estimates made herein
describe the occurrence of events along a continuous speciation process with particular
reference to nominal species. Changing the focal unit will shift the order of magnitude of
temporal estimates and may produce different biogeographical patterns. The inclusion of more
sequence data and the implementation of future analytical methods may also expand on the
work herein. Specifically, coalescent methods applied within the context of robust phylogenetic
frameworks may provide a more thorough interpretation of the evolutionary history of coral
124
reef fish species (Bowen et al., 2001; Nichols, 2001; Bowen et al., 2006b; van Herwerden et al.,
2006; Leray et al., 2010; Eytan et al., 2012; Harrington & Near, 2012; Lessios & Robertson,
2013). Experimental work and the incorporation of more detailed ecological data are also likely
to provide new evidence to evaluate the processes of divergence proposed herein (Munday et
al., 2004; Bolnick et al., 2006; Crow et al., 2010). It will be of great interest to discover the
ways through which sympatric sister-species arise and persist, and what role, if any, the Coral
Triangle plays in their concentrated coexistence.
125
References
Abellan, P. & Ribera, I. (2011) Geographic location and phylogeny are the main determinants
of the size of the geographical range in aquatic beetles. BMC Evolutionary Biology, 11,
344–358.
Ackiss, A.S., Pardede, S., Crandall, E.D., Ablan-Lagman, M.C.A., Ambariyanto, Romena, N.,
Barber, P.H. & Carpenter, K.E. (2013) Pronounced genetic structure in a highly mobile
coral reef fish, Caesio cuning, in the Coral Triangle. Marine Ecology Progress Series,
480, 185–197.
Alfaro, M.E., Santini, F. & Brock, C.D. (2007) Do reefs drive diversification in marine teleosts?
Evidence from the pufferfishes and their allies (Order Tetraodontiformes). Evolution,
61, 2104–2126.
Allen, G.R. (2008) Conservation hotspots of biodiversity and endemism for Indo-Pacific coral
reef fishes. Aquatic Conservation: Marine and Freshwater Ecosystems, 18, 541–556.
Allen, G.R., Steene, R. & Allen, M. (1998) A guide to Angelfishes and Butterflyfishes, 1st edn.
Odyssey Publishing/Tropical Reef Research, Perth.
Altman, S., Robinson, J., Pringle, J., Byers, J. & Wares, J. (2013) Edges and overlaps in
northwest Atlantic phylogeography. Diversity, 5, 263–275.
Alva-Campbell, Y., Floeter, S.R., Robertson, D.R., Bellwood, D.R. & Bernardi, G. (2010)
Molecular phylogenetics and evolution of Holacanthus angelfishes (Pomacanthidae).
Molecular Phylogenetics and Evolution, 56, 456–461.
Ashman, K.M., Bird, C.M. & Zepf, S.E. (1994) Detecting bimodality in astronomical datasets.
The Astronomical Journal, 108, 2348–2361.
Azuma, Y., Kumazawa, Y., Miya, M., Mabuchi, K. & Nishida, M. (2008) Mitogenomic
evaluation of the historical biogeography of cichlids toward reliable dating of teleostean
divergences. BMC Evolutionary Biology, 8, 215.
126
Bakus, G., Arthur, R., Ekaratne, S. & Jinendradasa, S.S. (2000) India and Sri Lanka. Coral
Reefs of the Indian Ocean: Their Ecology and Conservation (ed. by T.R. Mcclanahan,
C.R.C. Sheppard and D.O. Obura), pp. 295–316. Oxford University Press, New York.
Bannikov, A.F. & Sorbini, L. (1990) Eocoris bloti, a new genus and species of labrid fish
(Perciformes, Labroidei) from the Eocene of Monte Bolca, Italy. Studi e Recereche sui
Giacimenti Terziari di Bola, Museo Civico di Storia Naturale, Verona 6, 133–148.
Bannikov, A.F. & Tyler, J.C. (1995) Phylogenetic revision of the fish families Luvaridae and
Kushlukiidae (Acanthuroidei), with a new genus and two new species of Eocene
luvarids. Smithsonian Contributions to Paleobiology, 81, 1–45.
Barber, P.H. (2009) The challenge of understanding the Coral Triangle biodiversity hotspot.
Journal of Biogeography, 36, 1845–1846.
Barber, P.H. & Bellwood, D.R. (2005) Biodiversity hotspots: evolutionary origins of
biodiversity in wrasses (Halichoeres: Labridae) in the Indo-Pacific and new world
tropics. Molecular Phylogenetics and Evolution, 35, 235–253.
Barber, P.H., Palumbi, S.R., Erdmann, M.V. & Moosa, M.K. (2000) Biogeography: a marine
Wallace's line? Nature, 406, 692–693.
Barraclough, T.G. & Nee, S. (2001) Phylogenetics and speciation. Trends in Ecology and
Evolution, 16, 391–399.
Barraclough, T.G. & Vogler, A.P. (2000) Detecting the geographical pattern of speciation from
species level phylogenies. The American Naturalist, 155, 419–434.
Bay, L.K., Choat, H.J., van Herwerden, L. & Robertson, D.R. (2004) High genetic diversities
and complex genetic structure in an Indo-Pacific tropical reef fish (Chlorurus
sordidus): evidence of an unstable evolutionary past? Marine Biology, 144, 757–767.
Bellwood, D.R. (1997) Reef fish biogeography: habitat associations, fossils and phylogenies.
Proceedings of the 8th International Coral Reef Symposium (ed. by. H.A. Lessios and
I.G. Macintyre), pp. 379–384. Smithsonian Tropical Research Institute, Panama.
Bellwood, D.R. & Hughes, T.P. (2001) Regional-scale assembly rules and biodiversity of coral
reefs. Science, 292, 1532–1534.
127
Bellwood, D.R. & Meyer, C.P. (2009) Searching for heat in a marine biodiversity hotspot.
Journal of Biogeography, 36, 569–576.
Bellwood, D.R. & Wainwright, P.C. (2002) The history and biogeography of fishes on coral
reefs. Coral Reef Fishes: Dynamics and Diversity in a Complex System (ed. by P.F.
Sale), pp. 5–32. Academic Press, New York.
Bellwood, D.R., van Herwerden, L. & Konow, N. (2004) Evolution and biogeography of
marine angelfishes (Pisces: Pomacanthidae). Molecular Phylogenetics and Evolution,
33, 140–155.
Bellwood, D.R., Hughes, T.P., Connolly, S.R. & Tanner, J. (2005) Environmental and
geometric constraints on Indo-Pacific coral reef biodiversity. Ecology Letters, 8, 643–
651.
Bellwood, D.R., Klanten, S., Cowman, P.F., Pratchett, M.S., Konow, N. & van Herwerden, L.
(2010) Evolutionary history of the butterflyfishes (f: Chaetodontidae) and the rise of
coral feeding fishes. Journal of Evolutionary Biology, 23, 335–349.
Bellwood, D.R., Renema, W. & Rosen, B.R. (2012) Biodiversity hotspots, evolution and coral
reef biogeography: a review. Biotic Evolution and Environmental Change in Southeast
Asia (ed. by D.J. Gower, K.G. Johnson, J.E. Richardson, B.R. Rosen, L. Ruber and S.T.
Williams), pp. 216–245. Cambridge University Press, New York.
Bender, M.G., Pie, M.R., Rezende, E.L., Mouillot, D. & Floeter, S.R. (2013) Biogeographic,
historical and environmental influences on the taxonomic and functional structure of
Atlantic reef fish assemblages. Global Ecology and Biogeography, 22, 1173–1182.
Bernardi, G. & Lape, J. (2005) Tempo and mode of speciation in the Baja California disjunct
fish species Anisotremus davidsonii. Molecular Ecology, 14, 4085–4096.
Bernardi, G., Holbrook, S.J., Schmitt, R.J., Crane, N.L. & DeMartini, E. (2002) Species
boundaries, populations and colour morphs in the coral reef three-spot damselfish
(Dascyllus trimaculatus) species complex. Proceedings of the Royal Society B:
Biological Sciences, 269, 599–605.
128
Bernardi, G., Bucciarelli, G., Costagliola, D., Robertson, D. & Heiser, J. (2004) Evolution of
coral reef fish Thalassoma spp. (Labridae). 1. Molecular phylogeny and biogeography.
Marine Biology, 144, 369–375.
Bernardi, G., Alva-Campbell, Y.R., Gasparini, J.L. & Floeter, S.R. (2008) Molecular ecology,
speciation, and evolution of the reef fish genus Anisotremus. Molecular Phylogenetics
and Evolution, 48, 929–935.
Bird, C.E. (2011) Morphological and behavioral evidence for adaptive diversification of
sympatric Hawaiian limpets (Cellana spp.). Integrative and Comparative Biology, 51,
466–473.
Bird, C.E., Holland, B.S., Bowen, B.W. & Toonen, R.J. (2011) Diversification of sympatric
broadcast-spawning limpets (Cellana spp.) within the Hawaiian archipelago. Molecular
Ecology, 20, 2128–2141.
Bird, C.E., Fernandez-Silva, I., Skillings, D.J. & Toonen, R.J. (2012) Sympatric speciation in
the post “modern synthesis” era of evolutionary biology. Evolutionary Biology, 39,
158–180.
Biton, E., Gildor, H. & Peltier, W.R. (2008) Red Sea during the Last Glacial Maximum:
implications for sea level reconstruction. Paleoceanography, 23, PA1214.
Blum, S.D. (1989) Biogeography of the Chaetodontidae: an analysis of allopatry among closely
related species. Environmental Biology of Fishes, 25, 9–31.
Böhning-Gaese, K., Caprano, T., van Ewijk, K. & Veith, M. (2006) Range size: disentangling
current traits and phylogenetic and biogeographic factors. The American Naturalist,
167, 555–567.
Bolnick, D.I., Near, T.J. & Wainwright, P.C. (2006) Body size divergence promotes post-
zygotic reproductive isolation in centrarchids. Evolutionary Ecology Research, 8, 903–
913.
Bosworth, W., Huchon, P. & McClay, K. (2005) The Red Sea and Gulf of Aden basins. Journal
of African Earth Sciences, 43, 334–378.
129
Bowen, B.W., Bass, A.L., Rocha, L.A., Grant, W.S. & Robertson, D.R. (2001) Phylogeography
of the trumpetfishes (Aulostomus): ring species complex on a global scale. Evolution,
55, 1029–1039.
Bowen, B.W., Bass, A.L., Muss, A., Carlin, J. & Robertson, D.R. (2006a) Phylogeography of
two Atlantic squirrelfishes (Family Holocentridae): exploring links between pelagic
larval duration and population connectivity. Marine Biology, 149, 899–913.
Bowen, B.W., Muss, A., Rocha, L.A. & Grant, W.S. (2006b) Shallow mtDNA coalescence in
Atlantic pygmy angelfishes (Genus Centropyge) indicates a recent invasion from the
Indian Ocean. Journal of Heredity, 97, 1–12.
Bowen, B.W., Rocha, L.A., Toonen, R.J., Karl, S.A. & the Tobo Laboratory (2013) The origins
of tropical marine biodiversity. Trends in Ecology and Evolution, 28, 359–366.
Brandley, M.C., Wang, Y., Guo, X., Montes de Oca, A.N., Fería-Ortíz, M., Hikida, T. & Ota,
H. (2011) Accommodating heterogenous rates of evolution in molecular divergence
dating methods: an example using intercontinental dispersal of Plestiodon (Eumeces)
lizards. Systematic Biology, 60, 3–15.
Briggs, J.C. (1974) Marine Zoogeography. McGraw-Hill, New York.
Briggs, J.C. (1989) The historic biogeography of India: isolation or contact? Systematic
Biology, 38, 322–332.
Briggs, J.C. (1999) Coincident biogeographic patterns: Indo-West Pacific Ocean. Evolution, 53,
326–335.
Briggs, J.C. (2009) Diversity, endemism and evolution in the Coral Triangle. Journal of
Biogeography, 36, 2008–2010.
Briggs, J.C. & Bowen, B.W. (2012) A realignment of marine biogeographic provinces with
particular reference to fish distributions. Journal of Biogeography, 39, 12–30.
Briggs, J.C. & Bowen, B.W. (2013) Marine shelf habitat: biogeography and evolution. Journal
of Biogeography, 40, 1023–1035.
Brown, J.H. & Lomolino, M.V. (1998) Biogeography. Sinauer, Sunderland.
Brown, W.L. & Wilson, E.O. (1956) Character displacement. Systematic Biology, 5, 49–64.
130
Budd, A.F. & Pandolfi, J.M. (2010) Evolutionary novelty is concentrated at the edge of coral
species distributions. Science, 328, 1558–1561.
Butlin, R.K., Galindo, J. & Grahame, J.W. (2008) Sympatric, parapatric or allopatric: the most
important way to classify speciation? Philosophical Transactions of the Royal Society
B: Biological Sciences, 363, 2997–3007.
Butzin, M., Lohmann, G. & Bickert, T. (2011) Miocene ocean circulation inferred from marine
carbon cycle modeling combined with benthic isotope records. Paleoceanography, 26,
PA1203.
Carpenter, K.E., Barber, P.H., Crandall, E.D., Ablan-Lagman, M.C.A., Ambariyanto,
Mahardika, G.N., Manjaji-Matsumoto, B.M., Juinio-Meñez, M.A., Santos, M.D.,
Starger, C.J. & Toha, A.H.A. (2011) Comparative phylogeography of the Coral
Triangle and implications for marine management. Journal of Marine Biology, 2011,
14.
Castelin, M., Lambourdiere, J., Boisselier, M.-C., Lozouet, P., Couloux, A., Cruaud, C. &
Samadi, S. (2010) Hidden diversity and endemism on seamounts: focus on poorly
dispersive neogastropods. Biological Journal of the Linnean Society, 100, 420–438.
Chiba, S.N., Iwatsuki, Y., Yoshino, T. & Hanzawa, N. (2009) Comprehensive phylogeny of the
family Sparidae (Perciformes: Teleostei) inferred from mitochondrial gene analyses.
Genes & Genetic Systems, 84, 153–170.
Choat, J.H., Ayling, A.M. & Schiel, D.R. (1988) Temporal and spatial variation in an island
fish fauna. Journal of Experimental Marine Biology and Ecology, 121, 91–111.
Choat, J.H., Klanten, S.O., van Herwerden, L., Robertson, D.R. & Clements, K.D. (2012)
Pattern and process in the evolutionary history of parrotfishes (Family Labridae).
Biological Journal of the Linnean Society, 107, 529–557.
Chow, S. & Hazama, K. (1998) Universal PCR primers for S7 ribosomal protein gene introns in
fish. Molecular Ecology, 7, 1255–1256.
131
Clague, D.A. & Dalrymple, G.B. (1987) Tectonics, geochronology and origin of the Hawaiian-
Emperor volcanic chain. Volcanism in Hawaii (ed. by R.W. Decker, T.L. Wright and
P.H. Stauffer), pp. 1–54. US Government Printing Office, Washington DC.
Clift, P.D. & Blusztajn, J. (2005) Reorganization of the western Himalayan river system after
five million years ago. Nature, 438, 1001–1003.
Coates, A.G. & Obando, J.A. (1996) The geological evolution of the Central American isthmus.
Evolution and Environment in Tropical America (ed. by J.B.C. Jackson, A.F. Budd and
A.G. Coates), pp. 21–56. University Chicago Press, Chicago.
Coates, A.G., Jackson, J.B.C., Collins, L.S., Cronin, T.M., Dowsett, H.J., Bybell, L.M., Jung, P.
& Obando, J.A. (1992) Closure of the Isthmus of Panama: the near-shore marine record
of Costa Rica and western Panama. Geological Society of America Bulletin, 104, 814–
828.
Connolly, S.R., Bellwood, D.R. & Hughes, T.P. (2003) Indo-Pacific biodiversity of coral reefs:
deviations from a mid-domain model. Ecology, 84, 2178–2190.
Cooper, W.J., Smith, L.L. & Westneat, M.W. (2009) Exploring the radiation of a diverse reef
fish family: phylogenetics of the damselfishes (Pomacentridae), with new
classifications based on molecular analyses of all genera. Molecular Phylogenetics and
Evolution, 52, 1–16.
Cowman, P.F. & Bellwood, D.R. (2011) Coral reefs as drivers of cladogenesis: expanding coral
reefs, cryptic extinction events, and the development of biodiversity hotspots. Journal
of Evolutionary Biology, 24, 2543–2562.
Cowman, P.F. & Bellwood, D.R. (2013a) The historical biogeography of coral reef fishes:
global patterns of origination and dispersal. Journal of Biogeography, 40, 209–224.
Cowman, P.F. & Bellwood, D.R. (2013b) Vicariance across major marine biogeographic
barriers: temporal concordance and the relative intensity of hard versus soft barriers.
Proceedings of the Royal Society B: Biological Sciences, 280, 20131541.
132
Cowman, P.F., Bellwood, D.R. & van Herwerden, L. (2009) Dating the evolutionary origins of
wrasse lineages (Labridae) and the rise of trophic novelty on coral reefs. Molecular
Phylogenetics and Evolution, 52, 621–631.
Coyne, J.A. & Orr, H.A. (2004) Speciation. Sinauer Associates, Inc, Sunderland.
Craig, M.T., Hastings, P.A., Pondella, D.J., Ross Robertson, D. & Rosales-Casián, J.A. (2006)
Phylogeography of the flag cabrilla Epinephelus labriformis (Serranidae): implications
for the biogeography of the Tropical Eastern Pacific and the early stages of speciation
in a marine shore fish. Journal of Biogeography, 33, 969–979.
Craig, M.T., Eble, J.A., Bowen, B.W. & Robertson, D.R. (2007) High genetic connectivity
across the Indian and Pacific Oceans in the reef fish Myripristis berndti
(Holocentridae). Marine Ecology Progress Series, 334, 245–254.
Craig, M.T., Eble, J.A. & Bowen, B.W. (2010) Origins, ages and population histories:
comparative phylogeography of endemic Hawaiian butterflyfishes (genus Chaetodon).
Journal of Biogeography, 37, 2125–2136.
Crow, K.D., Kanamoto, Z. & Bernardi, G. (2004) Molecular phylogeny of the hexagrammid
fishes using a multi-locus approach. Molecular Phylogenetics and Evolution, 32, 986–
997.
Crow, K.D., Munehara, H. & Bernardi, G. (2010) Sympatric speciation in a genus of marine
reef fishes. Molecular Ecology, 19, 2089–2105.
Curray, J.R., Emmel, F.J. & Moore, D.G. (2003) The Bengal Fan: morphology, geometry,
stratigraphy, history and processes. Marine and Petroleum Geology, 19, 1191–1223.
Dalebout, M.L., Steel, D. & Baker, C.S. (2008) Phylogeny of the beaked whale genus
Mesoplodon (Ziphiidae: Cetacea) revealed by nuclear introns: implications for the
evolution of male tusks. Systematic Biology, 57, 857–875.
Darwin, C. (1859) On the Origin of Species by Means of Natural Selection or Preservation of
Favoured Races in the Struggle for Life. Murray, London.
DeBoer, T.S., Abdon Naguit, M.R., Erdmann, M.V., Ablan-Lagman, M.C.A., Ambariyanto,
Carpenter, K.E., Toha, A.H.A. & Barber, P.H. (2014) Concordance between
133
phylogeographic and biogeographic boundaries in the Coral Triangle: conservation
implications based on comparative analyses of multiple giant clam species. Bulletin of
Marine Science, 90, 277–300.
DiBattista, J.D., Wilcox, C., Craig, M.T., Rocha, L.A. & Bowen, B.W. (2011) Phylogeography
of the Pacific blueline surgeonfish, Acanthurus nigroris, reveals high genetic
connectivity and a cryptic endemic species in the Hawaiian archipelago. Journal of
Marine Biology, 2011, 17.
DiBattista, J.D., Waldrop, E., Bowen, B.W., Schultz, J.K., Gaither, M.R., Pyle, R.L. & Rocha,
L.A. (2012) Twisted sister species of pygmy angelfishes: discordance between
taxonomy, coloration, and phylogenetics. Coral Reefs, 31, 839–851.
DiBattista, J.D., Berumen, M.L., Gaither, M.R., Rocha, L.A., Eble, J.A., Choat, J.H., Craig,
M.T., Skillings, D.J. & Bowen, B.W. (2013) After continents divide: comparative
phylogeography of reef fishes from the Red Sea and Indian Ocean. Journal of
Biogeography, 40, 1170–1181.
Dickens, G.R. & Owen, R.M. (1999) The latest Miocene—Early Pliocene biogenic bloom: a
revised Indian Ocean perspective. Marine Geology, 161, 75–91.
Dietz, E.J. (1983) Permutation tests for association between two distance matrices. Systematic
Zoology, 32, 21–26.
Drew, J. & Barber, P.H. (2009) Sequential cladogenesis of the reef fish Pomacentrus
moluccensis (Pomacentridae) supports the peripheral origin of marine biodiversity in
the Indo-Australian archipelago. Molecular Phylogenetics and Evolution, 53, 335–339.
Drew, J.A. & Barber, P.H. (2012) Comparative phylogeography in Fijian coral reef fishes: a
multi-taxa approach towards marine reserve design. PLoS ONE, 7, e47710.
Drummond, A. & Rambaut, A. (2007) BEAST: Bayesian evolutionary analysis by sampling
trees. BMC Evolutionary Biology, 7, 214.
Drummond, A.J., Ho, S.Y.W., Phillips, M.J. & Rambaut, A. (2006) Relaxed phylogenetics and
dating with confidence. PLoS Biology, 4, 699–710.
Dulvy, N.K. (2006) Extinctions and threat in the sea. MarBEF Newsletter, Spring, 20–22.
134
Dulvy, N.K., Sadovy, Y. & Reynolds, J.D. (2003) Extinction vulnerability in marine
populations. Fish and Fisheries, 4, 25–64.
Eble, J.A., Toonen, R.J., Sorenson, L., Basch, L.V., Papastamatiou, Y.P. & Bowen, B.W.
(2011) Escaping paradise: larval export from Hawaii in an Indo-Pacific reef fish, the
yellow tang Zebrasoma flavescens. Marine Ecology Progress Series, 428, 245–258.
Ekman, S. (1953) Zoogeography of the Sea. Sidgwick and Jackson, London.
Exon, N.F., Quilty, P.G., Lafoy, Y., Crawford, A.J. & Auzende, J.M. (2004) Miocene volcanic
seamounts on northern Lord Howe Rise: lithology, age and origin. Australian Journal
of Earth Sciences, 51, 291–300.
Exon, N.F., Hill, P.J., Lafoy, Y., Heine, C. & Bernardel, G. (2006) Kenn Plateau off northeast
Australia: a continental fragment in the southwest Pacific jigsaw. Australian Journal of
Earth Sciences, 53, 541–564.
Eytan, R.I., Hastings, P.A., Holland, B.R. & Hellberg, M.E. (2012) Reconciling molecules and
morphology: molecular systematics and biogeography of Neotropical blennies
(Acanthemblemaria). Molecular Phylogenetics and Evolution, 62, 159–173.
Fauvelot, C., Bernardi, G. & Planes, S. (2003) Reductions in the mitochondrial DNA diversity
of coral reef fish provide evidence of population bottlenecks resulting from Holocene
sea-level change. Evolution, 57, 1571–1583.
Felsenstein, J. (1985) Phylogenies and the comparative method. The American Naturalist, 125,
1–15.
Fessler, J.L. & Westneat, M.W. (2007) Molecular phylogenetics of the butterflyfishes
(Chaetodontidae): taxonomy and biogeography of a global coral reef fish family.
Molecular Phylogenetics and Evolution, 45, 50–68.
Fitzpatrick, B.M. & Turelli, M. (2006) The geography of mammalian speciation: mixed signals
from phylogenies and range maps. Evolution, 60, 601–615.
Fitzpatrick, B.M., Fordyce, J.A. & Gavrilets, S. (2008) What, if anything, is sympatric
speciation? Journal of Evolutionary Biology, 21, 1452–1459.
135
Fitzpatrick, B.M., Fordyce, J.A. & Gavrilets, S. (2009) Pattern, process and geographic modes
of speciation. Journal of Evolutionary Biology, 22, 2342–2347.
Fitzpatrick, J.M., Carlon, D.B., Lippe, C. & Robertson, D.R. (2011) The West Pacific diversity
hotspot as a source or sink for new species? Population genetic insights from the Indo-
Pacific parrotfish Scarus rubroviolaceus. Molecular Ecology, 20, 219–234.
Fleischer, R.C., McIntosh, C.E. & Tarr, C.L. (1998) Evolution on a volcanic conveyor belt:
using phylogeographic reconstructions and K–Ar-based ages of the Hawaiian Islands to
estimate molecular evolutionary rates. Molecular Ecology, 7, 533–545.
Fleminger, A. (1986) The Pleistocene equatorial barrier between the Indian and Pacific Oceans
and a likely cause for Wallace’s line. Pelagic Biogeography (ed. by A.C. Pierrot-Bults,
S. Van Der Spoel, B.J. Aahuraner and R.K. Johnson), pp. 84–97. UNESCO Technical
Papers in Marine Science, Paris.
Floeter, S.R., Rocha, L.A., Robertson, D.R., Joyeux, J.C., Smith-Vaniz, W.F., Wirtz, P.,
Edwards, A.J., Barreiros, J.P., Ferreira, C.E.L., Gasparini, J.L., Brito, A., Falcón, J.M.,
Bowen, B.W. & Bernardi, G. (2008) Atlantic reef fish biogeography and evolution.
Journal of Biogeography, 35, 22–47.
Foote, M. (2007) Symmetric waxing and waning of marine invertebrate genera. Paleobiology,
33, 517–529.
Frey, M.A. (2010) The relative importance of geography and ecology in species diversification:
evidence from a tropical marine intertidal snail (Nerita). Journal of Biogeography, 37,
1515–1528.
Frey, M.A. & Vermeij, G.J. (2008) Molecular phylogenies and historical biogeography of a
circumtropical group of gastropods (Genus: Nerita): implications for regional diversity
patterns in the marine tropics. Molecular Phylogenetics and Evolution, 48, 1067–1086.
Gaither, M.R. & Rocha, L.A. (2013) Origins of species richness in the Indo-Malay-Philippine
biodiversity hotspot: evidence for the centre of overlap hypothesis. Journal of
Biogeography, 40, 1638–1648.
136
Gaither, M.R., Toonen, R.J., Robertson, D.R., Planes, S. & Bowen, B.W. (2010) Genetic
evaluation of marine biogeographical barriers: perspectives from two widespread Indo-
Pacific snappers (Lutjanus kasmira and Lutjanus fulvus). Journal of Biogeography, 37,
133–147.
Gaither, M.R., Bowen, B.W., Bordenave, T.-R., Rocha, L.A., Newman, S.J., Gomez, J.A., van
Herwerden, L. & Craig, M.T. (2011a) Phylogeography of the reef fish Cephalopholis
argus (Epinephelidae) indicates Pleistocene isolation across the Indo-Pacific barrier
with contemporary overlap in the Coral Triangle. BMC Evolutionary Biology, 11, 189.
Gaither, M.R., Jones, S.A., Kelley, C., Newman, S.J., Sorenson, L. & Bowen, B.W. (2011b)
High connectivity in the deepwater snapper Pristipomoides filamentosus (Lutjanidae)
across the Indo-Pacific with isolation of the Hawaiian Archipelago. PLoS One, 6,
e28913.
Gaither, M., Schultz, J.K., Bellwood, D.R., Pyle, R.L., DiBattista, J.D., Rocha, L.A. & Bowen,
B.W. (2014) Evolution of pygmy angelfishes: recent divergences, introgression, and the
usefulness of color in taxonomy. Molecular Phylogenetics and Evolution, 74, 38–47.
Garrigan, D., Kingan, S.B., Geneva, A.J., Andolfatto, P., Clark, A.G., Thornton, K.R. &
Presgraves, D.C. (2012) Genome sequencing reveals complex speciation in the
Drosophila simulans clade. Genome Research, 22, 1499–1511.
Gaston, K.J. (1998) Species-range size distributions: products of speciation, extinction and
transformation. Philosophical Transactions of the Royal Society B: Biological Sciences,
353, 219–230.
Gaston, K.J. (2003) The structure and dynamics of geographic range. Oxford Series in Ecology
and Evolution. Oxford University Press, Oxford.
Gaston, K.J. & Blackburn, T.M. (1997) Age, area and avian diversification. Biological Journal
of the Linnean Society, 62, 239–253.
Gavrilets, S. (2000) Waiting time to parapatric speciation. Proceedings of the Royal Society B:
Biological Sciences, 267, 2483–2492.
137
Gernhard, T. (2008) The conditioned reconstructed process. Journal of Theoretical Biology,
253, 769–778.
GRASS Development Team (2011) Geographic Resources Analysis Support System (GRASS).
Open Source Geospatial Foundation Project. Available at: http://grass.osgeo.org.
Halas, D. & Winterbottom, R. (2009) A phylogenetic test of multiple proposals for the origins
of the East Indies coral reef biota. Journal of Biogeography, 36, 1847–1860.
Hall, I.R., McCave, I.N., Shackleton, N.J., Weedon, G.P. & Harris, S.E. (2001) Intensified deep
Pacific inflow and ventilation in Pleistocene glacial times. Nature, 412, 809–812.
Hall, R. (1998) The plate tectonics of Cenozoic SE Asia and the distribution of land and sea.
Biogeography and Geological Evolution of SE Asia (ed. by R. Hall and J.D. Holloway),
pp. 99–131. Backbuys Publishers, Leiden.
Hardin, G. (1960) The competitive exclusion principle. Science, 131, 1292–1297.
Harrington, R.C. & Near, T.J. (2012) Phylogenetic and coalescent strategies of species
delimitation in snubnose darters (Percidae: Etheostoma). Systematic Biology, 61, 63–
79.
Harvey, P.H., May, R.M. & Nee, S. (1994) Phylogenies without fossils. Evolution, 48, 523–
529.
Hastings, P.A. (1990) Phylogenetic relationships of tube blennies of the genus
Acanthemblemaria (Pisces: Blennioidei). Bulletin of Marine Science, 47, 725–738.
Hastings, P.A. & Springer, V.G. (1994) Review of Stathmonotus, with redefinition and
phylogenetic analysis of the Chaenopsidae (Teleostei: Blennioidei). Smithsonian
Contributions to Zoology, 558, 1–48.
Hawkins, J.P., Roberts, C.M. & Clark, V. (2000) The threatened status of restricted-range coral
reef fish species. Animal Conservation, 3, 81–88.
Heads, M. (2009) Inferring biogeographic history from molecular phylogenies. Biological
Journal of the Linnean Society, 98, 757–774.
Heck, K.L. & McCoy, E.D. (1978) Long-distance dispersal and the reef-building corals of the
Eastern Pacific. Marine Biology, 48, 349–356.
138
Hellberg, M.E. (2009) Gene flow and isolation among populations of marine animals. Annual
Review of Ecology, Evolution, and Systematics, 40, 291–310.
Hemleben, C., Meischner, D., Zahn, R., Almogi-Labin, A., Erlenkeuser, H. & Hiller, B. (1996)
Three hundred eighty thousand year long stable isotope and faunal records from the
Red Sea: influence of global sea level change on hydrography. Paleoceanography, 11,
147–156.
Hobbs, J.-P. & van Herwerden, L. (2010) Evolutionary novelty at ranges for coral reef
organisms. Science E-Letters.
Hobbs, J.-P.A., Frisch, A.J., Allen, G.R. & van Herwerden, L. (2009) Marine hybrid hotspot at
Indo-Pacific biogeographic border. Biology Letters, 5, 258–261.
Hoeksema, B.W. (2007) Delineation of the Indo-Malayan centre of maximum marine
biodiversity: the Coral Triangle. Biogeography, Time, and Place: Distributions,
Barriers, and Islands (ed. by W. Renema), pp. 117–178. Springer Netherlands,
Dordrecht.
Hoorn, C., Guerrero, J., Sarmiento, G.A. & Lorente, M.A. (1995) Andean tectonics as a cause
for changing drainage patterns in Miocene northern South America. Geology, 23, 237–
240.
Horne, J.B., van Herwerden, L., Choat, J.H. & Robertson, D.R. (2008) High population
connectivity across the Indo-Pacific: congruent lack of phylogeographic structure in
three reef fish congeners. Molecular Phylogenetics and Evolution, 49, 629–638.
Hoskin, C.J., Tonione, M., Higgie, M., Jason, B.M., Williams, S.E., VanDerWal, J. & Moritz,
C. (2011) Persistence in peripheral refugia promotes phenotypic divergence and
speciation in a rainforest frog. The American Naturalist, 178, 561–578.
Hourigan, T.F. & Reese, E.S. (1987) Mid-ocean isolation and the evolution of Hawaiian reef
fishes. Trends in Ecology & Evolution, 2, 187–191.
Hubbell, S.P. (2001) The Unified Neutral Theory of Biodiversity and Biogeography. Princeton
University Press, Princeton.
139
Hug, L.A. & Roger, A.J. (2007) The impact of fossils and taxon sampling on ancient molecular
dating analyses. Molecular Biology and Evolution, 24, 1889–1897.
Hughes, T.P., Bellwood, D.R. & Connolly, S.R. (2002) Biodiversity hotspots, centres of
endemicity, and the conservation of coral reefs. Ecology Letters, 5, 775–784.
Hughes, T.P., Baird, A.H., Bellwood, D.R., Card, M., Connolly, S.R., Folke, C., Grosberg, R.,
Hoegh-Guldberg, O., Jackson, J.B.C., Kleypas, J., Lough, J.M., Marshall, P., Nystrom,
M., Palumbi, S.R., Pandolfi, J.M., Rosen, B. & Roughgarden, J. (2003) Climate change,
human impacts, and the resilience of coral reefs. Science, 301, 929–933.
Hurley, I.A., Mueller, R.L., Dunn, K.A., Schmidt, E.J., Friedman, M., Ho, R.K., Prince, V.E.,
Yang, Z., Thomas, M.G. & Coates, M.I. (2007) A new time-scale for ray-finned fish
evolution. Proceedings of the Royal Society B: Biological Sciences, 274, 489–498.
Hurvich, C.M. & Tsai, C.-L. (1989) Regression and time series model selection in small
samples. Biometrika, 76, 297–307.
Ingram, T. (2011) Speciation along a depth gradient in a marine adaptive radiation. Proceedings
of the Royal Society B: Biological Sciences, 278, 613–618.
IUCN (2011) IUCN Red List of Threatened Species. Version 2011.2. International Union for
Conservation of Nature, Gland. Available at: http://www.iucnredlist.org/.
Jablonski, D. (1987) Heritability at the species level: analysis of geographic ranges of
Cretaceous mollusks. Science, 238, 360–363.
Jablonski, D. & Hunt, G. (2006) Larval ecology, geographic range, and species survivorship in
Cretaceous mollusks: organismic versus species-level explanations. The American
Naturalist, 168, 556–564.
Jokiel, P.L. (1987) Ecology, biogeography and evolution of corals in Hawaii. Trends in Ecology
& Evolution, 2, 179–182.
Jones, G.P., Munday, P.L. & Julian Caley, M. (2002) Rarity in coral reef fish communities.
Coral Reef Fishes (ed. by P.F. Sale), pp. 81–101. Academic Press, San Diego.
Jones, K.E., Sechrest, W. & Gittleman, J.L. (2005) Age and area revisited: identifying global
patterns and implications for conservation. Phylogeny and Conservation (ed. by A.
140
Purvis, J.L. Gittleman and T. Brooks), pp. 142–165. Cambridge University Press,
Cambridge.
Kaneps, A.G. (1979) Gulf Stream: velocity fluctuations during the Late Cenozoic. Science, 204,
279–301.
Kay, E.A. & Palumbi, S.R. (1987) Endemism and evolution in Hawaiian marine invertebrates.
Trends in Ecology and Evolution, 2, 183–186.
Kazancıoğlu, E., Near, T.J., Hanel, R. & Wainwright, P.C. (2009) Influence of sexual selection
and feeding functional morphology on diversification rate of parrotfishes (Scaridae).
Proceedings of the Royal Society B: Biological Sciences, 276, 3439–3446.
Kirkendale, L.A. & Meyer, C.P. (2004) Phylogeography of the Patelloida profunda group
(Gastropoda: Lottidae): diversification in a dispersal-driven marine system. Molecular
Ecology, 13, 2749–2762.
Klanten, S.O., van Herwerden, L., Choat, J.H. & Blair, D. (2004) Patterns of lineage
diversification in the genus Naso (Acanthuridae). Molecular Phylogenetics and
Evolution, 32, 221–235.
Klanten, S.O., Choat, J.H. & van Herwerden, L. (2007) Extreme genetic diversity and temporal
rather than spatial partitioning in a widely distributed coral reef fish. Marine Biology,
150, 659–670.
Krug, P.J. (2011) Patterns of speciation in marine gastropods: a review of the phylogenetic
evidence for localized radiations in the sea. American Malacological Bulletin, 29, 169–
186.
Kuiter, R.H. (2010) Labridae Fishes: Wrasses, 1st edn. Aquatic Photographics, Seaford.
Kulbicki, M., Parravicini, V., Bellwood, D.R., Arias-Gonzàlez, E., Chabanet, P., Floeter, S.R.,
Friedlander, A., McPherson, J., Myers, R.E., Vigliola, L. & Mouillot, D. (2013) Global
biogeography of reef fishes: a hierarchical quantitative delineation of regions. PLoS
One, 8, e81847.
Kumar, S. (2005) Molecular clocks: four decades of evolution. Nature Reviews Genetics, 6,
654–662.
141
Kuriiwa, K., Hanzawa, N., Yoshino, T., Kimura, S. & Nishida, M. (2007) Phylogenetic
relationships and natural hybridization in rabbitfishes (Teleostei: Siganidae) inferred
from mitochondrial and nuclear DNA analyses. Molecular Phylogenetics and
Evolution, 45, 69–80.
Lacson, J.M. & Nelson, S.G. (1993) Genetic distances among fishes of the genus Siganus
(Siganidae) from the Western Pacific Ocean. Marine Biology, 116, 187–192.
Ladd, H.S. (1960) Origin of the Pacific Island molluscan fauna. American Journal of Science,
258A, 137–150.
Lanfear, R., Calcott, B., Ho, S.Y.W. & Guindon, S. (2012) PARTITIONFINDER: combined
selection of partitioning schemes and substitution models for phylogenetic analyses.
Molecular Biology and Evolution, 29, 1695–1701.
Leray, M., Beldade, R., Holbrook, S.J., Schmitt, R.J., Planes, S. & Bernardi, G. (2010)
Allopatric divergence and speciation in coral reef fish: the three-spot dascyllus,
Dascyllus trimaculatus, species complex. Evolution, 64, 1218–1230.
Lessios, H.A. (2008) The great American schism: divergence of marine organisms after the rise
of the Central American Isthmus. Annual Review of Ecology, Evolution and
Systematics, 39, 63–91.
Lessios, H.A. & Robertson, D.R. (2006) Crossing the impassable: genetic connections in 20
reef fishes across the eastern Pacific barrier. Proceedings of the Royal Society B:
Biological Sciences, 273, 2201–2208.
Lessios, H.A. & Robertson, D.R. (2013) Speciation on a round planet: phylogeography of the
goatfish genus Mulloidichthys. Journal of Biogeography, 40, 2373–2384.
Lessios, H.A., Kessing, B.D. & Robertson, D.R. (1998) Massive gene flow across the world's
most potent marine biogeographic barrier. Proceedings of the Royal Society B:
Biological Sciences, 265, 583–588.
Lessios, H.A., Kessing, B.D. & Pearse, J.S. (2001) Population structure and speciation in
tropical seas: global phylogeography of the sea urchin Diadema. Evolution, 55, 955–
975.
142
Lin, H.-C. & Hastings, P.A. (2013) Phylogeny and biogeography of a shallow water fish clade
(Teleostei: Blenniiformes). BMC Evolutionary Biology, 13, 210.
Lisiecki, L.E. & Raymo, M.E. (2005) A Pliocene-Pleistocene stack of 57 globally distributed
benthic δ18O records. Paleoceanography, 20, 1–17.
Longhurst, A.R. (1998) Ecological Geography of the Sea. Academic Press, San Diego.
Losos, J.B. & Glor, R.E. (2003) Phylogenetic comparative methods and the geography of
speciation. Trends in Ecology and Evolution, 18, 220–227.
Luiz, O.J., Madin, J.S., Robertson, D.R., Rocha, L.A., Wirtz, P. & Floeter, S.R. (2012)
Ecological traits influencing range expansion across large oceanic dispersal barriers:
insights from tropical Atlantic reef fishes. Proceedings of the Royal Society B:
Biological Sciences, 279, 1033–1040.
Lyle, M., Dadey, K.A. & Farrell, J.W. (1995) The late Miocene (11–8 Ma) eastern Pacific
carbonate crash: evidence for reorganization of deep-water circulation by the closure of
the Panama gateway. Proceedings of the Ocean Drilling Program, Scientific Results,
138, 821–838.
Lynch, J.D. (1989) The gauge of speciation: on the frequencies of modes of speciation.
Speciation and Its Consequences (ed. by D. Otte and J.A. Endler), pp. 527–553. Sinauer
Associates, Sunderland.
Malay, M.C.M.D. & Paulay, G. (2010) Peripatric speciation drives diversification and
distributional pattern of reef hermit crabs (Decapoda: Diogenidae: Calinus). Evolution,
64, 634–662.
Mallet, J., Meyer, A., Nosil, P. & Feder, J.L. (2009) Space, sympatry and speciation. Journal of
Evolutionary Biology, 22, 2332–2341.
Mantooth, S.J. & Riddle, B.R. (2011) Molecular biogeography: the intersection between
geographic and molecular variation. Geography Compass, 5, 1–20.
Mayr, E. (1942) Systematics and the Origin of Species. Columbia University Press, New York.
Mayr, E. (1944) Wallace's Line in the light of recent zoogeographic studies. The Quarterly
Review of Biology, 19, 1–14.
143
Mayr, E. (1954) Geographic speciation in tropical echinoids. Evolution, 8, 1–18.
Mayr, E. (1963) Animal Species and Evolution. Belknap Press, Cambridge.
McCafferty, S., Bermingham, E., Quenouille, B., Planes, S., Hoelzer, G. & Asoh, K. (2002)
Historical biogeography and molecular systematics of the Indo-Pacific genus Dascyllus
(Teleostei: Pomacentridae). Molecular Ecology, 11, 1377–1392.
McDougall, I. & Duncan, R.A. (1988) Age progressive volcanism in the Tasmantid Seamounts.
Earth and Planetary Science Letters, 89, 207–220.
McDougall, I., Embleton, B.J.J. & Stone, D.B. (1981) Origin and evolution of Lord Howe
Island, Southwest Pacific Ocean. Journal of the Geological Society of Australia, 28,
155–176.
McManus, J.W. (1985) Marine speciation, tectonics and sea-level changes in Southeast Asia.
Proceedings of the 5th International Coral Reef Congress, 4, 133–138.
McMillan, W.O. & Palumbi, S.R. (1995) Concordant evolutionary patterns among Indo-West
Pacific butterflyfishes. Proceedings of the Royal Society B: Biological Sciences, 260,
229–236.
McMillan, W.O., Weigt, L.A. & Palumbi, S.R. (1999) Color pattern evolution, assortative
mating, and genetic differentiation in brightly colored butterflyfishes (Chaetodontidae).
Evolution, 53, 247–260.
Meyer, C.P., Geller, J.B. & Paulay, G. (2005) Fine scale endemism on coral reefs: archipelagic
differentiation in turbinid gastropods. Evolution, 59, 113–125.
Miller, A.I. (1997) A new look at age and area: the geographic and environmental expansion of
genera during the Ordovician Radiation. Paleobiology, 23, 410–419.
Miller, M.A., Holder, M.T., Vos, R., Midford, P.E., Liebowitz, T., Chan, L., Hoover, P. &
Warnow, T. (2010) The CIPRES Science Gateway V. 3.1. Available at:
http://www.phylo.org/sub_sections/portal.
Milne, R.I. (2009) Effects of taxon sampling on molecular dating for within-genus divergence
events, when deep fossils are used for calibration. Journal of Systematics and
Evolution, 47, 383–401.
144
Montanari, S.R., van Herwerden, L., Pratchett, M.S., Hobbs, J.-P.A. & Fugedi, A. (2012) Reef
fish hybridization: lessons learnt from butterflyfishes (genus Chaetodon). Ecology and
Evolution, 2, 310–328.
Mora, C., Treml, E.A., Roberts, J., Crosby, K., Roy, D. & Tittensor, D.P. (2012) High
connectivity among habitats precludes the relationship between dispersal and range size
in tropical reef fishes. Ecography, 35, 89–96.
Morlon, H. (2014) Phylogenetic approaches for studying diversification. Ecology Letters, 17,
508–525.
Munday, P.L. (2004) Habitat loss, resource specialization, and extinction on coral reefs. Global
Change Biology, 10, 1642–1647.
Munday, P.L., van Herwerden, L. & Dudgeon, C.L. (2004) Evidence for sympatric speciation
by host shift in the sea. Current biology, 14, 1498–1504.
Muss, A., Robertson, D.R., Stepien, C.A., Wirtz, P. & Bowen, B.W. (2001) Phylogeography of
Ophioblennius: the role of ocean currents and geography in reef fish evolution.
Evolution, 55, 561–572.
Near, T.J., Eytan, R.I., Dornburg, A., Kuhn, K.L., Moore, J.A., Davis, M.P., Wainwright, P.C.,
Friedman, M. & Smith, W.L. (2012) Resolution of ray-finned fish phylogeny and
timing of diversification. Proceedings of the National Academy of Sciences USA, 109,
13698–13703.
Near, T.J., Dornburg, A., Eytan, R.I., Keck, B.P., Smith, W.L., Kuhn, K.L., Moore, J.A., Price,
S.A., Burbrink, F.T., Friedman, M. & Wainwright, P.C. (2013) Phylogeny and tempo of
diversification in the superradiation of spiny-rayed fishes. Proceedings of the National
Academy of Sciences USA, 110, 12738–12743.
Nei, M. (1987) Molecular evolutionary genetics, First edn. Columbia University Press, New
York.
Nichols, R. (2001) Gene trees and species trees are not the same. Trends in Ecology and
Evolution, 16, 358–364.
145
O'Dea, A., Jackson, J.B.C., Fortunato, H., Smith, J.T., D'Croz, L., Johnson, K.G. & Todd, J.A.
(2007) Environmental change preceded Caribbean extinction by 2 million years.
Proceedings of the National Academy of Sciences USA, 104, 5501–5506.
Palumbi, S.R. (1997) Molecular biogeography of the Pacific. Coral Reefs, 16, S47–S52.
Pandolfi, J.M. (1992) Successive isolation rather than evolutionary centres for the origination of
Indo-Pacific reef corals. Journal of Biogeography, 19, 593–609.
Paul, J.R. & Tonsor, S.J. (2008) Explaining geographic range size by species age: a test using
Neotropical Piper species. Tropical forest community ecology (ed. by W.P. Carson & S.
Schnitzer), pp 46–62. Blackwell, London.
Paul, J.R., Morton, C., Taylor, C.M. & Tonsor, S.J. (2009) Evolutionary time for dispersal
limits the extent but not the occupancy of species' potential ranges in the tropical plant
genus Psychotria (Rubiaceae). The American Naturalist, 173, 188–199.
Paulay, G. (1990) Effects of Late Cenozoic sea-level fluctuations on the bivalve faunas of
tropical oceanic islands. Paleobiology, 16, 415–434.
Paulay, G. & Meyer, C. (2002) Diversification in the tropical Pacific: comparisons between
marine and terrestrial systems and the importance of founder speciation. Integrative and
Comparative Biology, 42, 922–934.
Phillimore, A.B., Orme, C.D.L., Thomas, G.H., Blackburn, T.M., Bennett, P.M., Gaston, K.J. &
Owens, I.P.F. (2008) Sympatric speciation in birds is rare: insights from range data and
simulations. The American Naturalist, 171, 646–657.
Pigot, A.L., Owens, I.P.F. & Orme, C.D.L. (2012) Speciation and extinction drive the
appearance of directional range size evolution in phylogenies and the fossil record.
PLoS Biology, 10, e1001260.
Pimm, S.L. & Raven, P. (2000) Biodiversity: extinction by numbers. Nature, 403, 843–845.
Posada, D. (2008) JMODELTEST: phylogenetic model averaging. Molecular Biology and
Evolution, 25, 1253–1256.
146
Posadas, P., Grossi, M.A. & Ortiz-Jaureguizar, E. (2013) Where is historical biogeography
going? The evolution of the discipline in the first decade of the 21st century. Progress
in Physical Geography, 37, 377–396.
Potts, D.C. (1985) Sea-level fluctuations and speciation in Scleractinia. Proceedings of the Fifth
International Coral Reef Congress, Tahiti, 4, 127–132.
Priest, M.A., Halford, A.R. & McIlwain, J.L. (2012) Evidence of stable genetic structure across
a remote island archipelago through self-recruitment in a widely dispersed coral reef
fish. Ecology and Evolution, 2, 3195–3213.
Przeslawski, R., Williams, A., Nichol, S.L., Hughes, M.G., Anderson, T.J. & Althaus, F. (2011)
Biogeography of the Lord Howe Rise region, Tasman Sea. Deep Sea Research Part II:
Topical Studies in Oceanography, 58, 959–969.
Puebla, O. (2009) Ecological speciation in marine v. freshwater fishes. Journal of Fish Biology,
75, 960–996.
QGIS Development Team (2012) QGIS Geographic Information System. Open Source Geospatial
Foundation Project. Available at: http://qgis.osgeo.org.
Queiroz, A.d. (2005) The resurrection of oceanic dispersal in historical biogeography. Trends in
Ecology and Evolution, 20, 68–73.
Quenouille, B., Hubert, N., Bermingham, E. & Planes, S. (2011) Speciation in tropical seas:
allopatry followed by range change. Molecular Phylogenetics and Evolution, 58, 564–
552.
Quilty, P.G. (1993) Tasmantid and Lord Howe seamounts: biostratigraphy and
palaeoceanographic significance. Alcheringa: An Australasian Journal of
Palaeontology, 17, 27–53.
R Development Core Team (2011) R: A Language and Environment for Statistical Computing.
R Foundation for Statistical Computing, Vienna. Available at: http://www.r-
project.org/.
Rabosky, D.L. & Lovette, I.J. (2008) Explosive evolutionary radiations: decreasing speciation
or increasing extinction through time? Evolution, 62, 1866–1875.
147
Rambaut, A. & Drummond, A.J. (2007) TRACER v1.4. Available at:
http://beast.bio.ed.ac.uk/Tracer.
Randall, J.E. (1972) A revision of the labrid fish genus Anampses. Micronesica, 8, 151–195.
Randall, J.E. (1986) Labridae. Smiths’ Sea Fishes (ed. by M.M. Smith and P.C. Heemstra), pp.
683–706. Springer-Verlag, Berlin.
Randall, J.E. (1998) Zoogeography of shore fishes of the Indo-Pacific region. Zoological
Studies, 37, 227–268.
Randall, J.E. (Unpublished) Randall's tank photos. Collection of 10,000 large-format photos
(slides) of dead fishes. Available at: http://fishbase.org/.
Raynal, J.M., Crandall, E.D., Barber, P.H., Mahardika, G.N., Lagman, M.C. & Carpenter, K.E.
(2014) Basin isolation and oceanographic features influencing lineage divergence in the
humbug damselfish (Dascyllus aruanus) in the Coral Triangle. Bulletin of Marine
Science, 90, 513–532.
Rea, D.K. & Bloomstine, M.K. (1986) Neogene history of the South Pacific tradewinds:
evidence for hemispherical asymmetry of atmospheric circulation. Palaeogeography,
Palaeoclimatology, Palaeoecology, 55, 55–64.
Read, C.I., Bellwood, D.R. & van Herwerden, L. (2006) Ancient origins of Indo-Pacific coral
reef fish biodiversity: a case study of the leopard wrasses (Labridae:
Macropharyngodon). Molecular Phylogenetics and Evolution, 38, 808–819.
Reaka, M.L., Rodgers, P.J. & Kudla, A.U. (2008) Patterns of biodiversity and endemism on
Indo-West Pacific coral reefs. Proceedings of the National Academy of Sciences USA,
105, 11474–11481.
Reece, J.S., Bowen, B.W., Joshi, K., Goz, V. & Larson, A. (2010) Phylogeography of two
moray eels indicates high dispersal throughout the Indo-Pacific. Journal of Heredity,
101, 391–402.
Reid, D.G., Dyal, P. & Williams, S.T. (2010) Global diversification of mangrove fauna: a
molecular phylogeny of Littoraria (Gastropoda: Littorinidae). Molecular Phylogenetics
and Evolution, 55, 185–201.
148
Reid, D.G., Lal, K., Mackenzie-Dodds, J., Kaligis, F., Littlewood, D.T.J. & Williams, S.T.
(2006) Comparative phylogeography and species boundaries in Echinolittorina snails in
the central Indo-West Pacific. Journal of Biogeography, 33, 990–1006.
Renema, W., Bellwood, D.R., Braga, J.C., Bromfield, K., Hall, R., Johnson, K.G., Lunt, P.,
Meyer, C.P., McMonagle, L.B., Morley, R.J., O'Dea, A., Todd, J.A., Wesselingh, F.P.,
Wilson, M.E.J. & Pandolfi, J.M. (2008) Hopping hotspots: global shifts in marine
biodiversity. Science, 321, 654–657.
Ricklefs, R.E. (2007) Estimating diversification rates from phylogenetic information. Trends in
Ecology and Evolution, 22, 601–610.
Ricklefs, R.E. & Bermingham, E. (2002) The concept of the taxon cycle in biogeography.
Global Ecology and Biogeography, 11, 353–361.
Ricklefs, R.E. & Cox, G.W. (1972) Taxon cycles in the West Indian avifauna. The American
Naturalist, 106, 195–219.
Riddle, B.R., Dawson, M.N., Hadly, E.A., Hafner, D.J., Hickerson, M.J., Mantooth, S.J. &
Yoder, A.D. (2008) The role of molecular genetics in sculpting the future of integrative
biogeography. Progress in Physical Geography, 32, 173–202.
Roberts, C.M., McClean, C.J., Veron, J.E.N., Hawkins, J.P., Allen, G.R., McAllister, D.E.,
Mittermeier, C.G., Schueler, F.W., Spalding, M., Wells, F., Vynne, C. & Werner, T.B.
(2002) Marine biodiversity hotspots and conservation priorities for tropical reefs.
Science, 295, 1280–1284.
Robertson, D.R., Karg, F., Moura, R.L.d., Victor, B.C. & Bernardi, G. (2006) Mechanisms of
speciation and faunal enrichment in Atlantic parrotfishes. Molecular Phylogenetics and
Evolution, 40, 795–807.
Rocha, L.A. (2003) Patterns of distribution and processes of speciation in Brazilian reef fishes.
Journal of Biogeography, 30, 1161–1171.
Rocha, L.A. & Bowen, B.W. (2008) Speciation in coral-reef fishes. Journal of Fish Biology, 72,
1101–1121.
149
Rocha, L.A., Bass, A.L., Robertson, D.R. & Bowen, B.W. (2002) Adult habitat preferences,
larval dispersal, and the comparative phylogeography of three Atlantic surgeonfishes
(Teleostei: Acanthuridae). Molecular Ecology, 11, 243–251.
Rocha, L.A., Robertson, D.R., Roman, J. & Bowen, B.W. (2005) Ecological speciation in
tropical reef fishes. Proceedings of the Royal Society B: Biological Sciences, 272, 573–
579.
Rocha, L.A., Craig, M.T. & Bowen, B.W. (2007) Phylogeography and the conservation of coral
reef fishes. Coral Reefs, 26, 513–513.
Rocha, L.A., Lindeman, K.C., Rocha, C.R. & Lessios, H.A. (2008) Historical biogeography and
speciation in the reef fish genus Haemulon (Teleostei: Haemulidae). Molecular
Phylogenetics and Evolution, 48, 918–928.
Ronquist, F. & Huelsenbeck, J.P. (2003) MRBAYES 3: Bayesian phylogenetic inference under
mixed models. Bioinformatics, 19, 1572–1574.
Rooney, J.J., Wessel, P., Hoeke, R., Weiss, J., Baker, J., Parrish, F., Fletcher, C.H., Chojnacki,
J., Garcia, M., Brainard, R. & Vroom, P. (2008) Geology and geomorphology of coral
reefs in the northwestern Hawaiian Islands. Coral Reefs of the USA (ed. by B.M. Riegl
and R.E. Dodge), pp. 519–571. Springer.
Rosen, B.R. (1984) Reef coral biogeography and climate through the Late Cainozoic: just
islands in the sun or a critical pattern of islands? Fossils and Climate (ed. by P.J.
Brenchley), pp. 201–262. Wiley, Chichester.
Rotondo, G.M., Springer, V.G., Scott, G.A.J. & Schlanger, S.O. (1981) Plate movement and
island integration – a possible mechanism in the formation of endemic biotas, with
special reference to the Hawaiian Islands. Systematic Zoology, 30, 12–21.
Russell, B.C. & Craig, M.T. (2013) Anampses viridis Valenciennes 1840 (Pisces: Labridae) – a
case of taxonomic confusion and mistaken extinction. Zootaxa, 3722, 083–091.
Samadi, S., Bottan, L., Macpherson, E., De Forges, B. & Boisselier, M.-C. (2006) Seamount
endemism questioned by the geographic distribution and population genetic structure of
marine invertebrates. Marine Biology, 149, 1463–1475.
150
Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2d
edn. Cold Spring Harbor Laboratory Press Cold Spring Harbor, New York.
Santini, F., Harmon, L., Carnevale, G. & Alfaro, M. (2009) Did genome duplication drive the
origin of teleosts? A comparative study of diversification in ray-finned fishes. BMC
Evolutionary Biology, 9, 194.
Santini, F., Miglietta, M. & Faucci, A. (2012) Speciation: where are we now? An introduction
to a special issue on speciation. Evolutionary Biology, 39, 141–147.
Schwarz, G. (1978) Estimating the dimension of a model. The Annals of Statistics, 6, 461–464.
Seddon, N., Botero, C.A., Tobias, J.A., Dunn, P.O., MacGregor, H.E.A., Rubenstein, D.R., Uy,
J.A.C., Weir, J.T., Whittingham, L.A. & Safran, R.J. (2013) Sexual selection
accelerates signal evolution during speciation in birds. Proceedings of the Royal Society
B: Biological Sciences, 280, 20131065.
Siddall, M., Rohling, E.J., Almogi-Labin, A., Hemleben, C., Meischner, D., Schmelzer, I. &
Smeed, D.A. (2003) Sea-level fluctuations during the last glacial cycle. Nature, 423,
853–858.
Singh, I.B. (2008) The Ganga river. Large Rivers: Geomorphology and Management (ed. by A.
Gupta), pp. 347–371. Wiley, West Sussex.
Skillings, D.J., Bird, C.E. & Toonen, R.J. (2011) Gateways to Hawai'i: genetic population
structure of the tropical sea cucumber Holothuria atra. Journal of Marine Biology,
2011, 16.
Sorenson, L., Santini, F., Carnevale, G. & Alfaro, M.E. (2013) A multi-locus timetree of
surgeonfishes (Acanthuridae, Percomorpha), with revised family taxonomy. Molecular
Phylogenetics and Evolution, 68, 150–160.
Spalding, M.D., Fox, H.E., Allen, G.R., Davidson, N., Ferdaña, Z.A., Finlayson, M., Halpern,
B.S., Jorge, M.A., Lombana, A., Lourie, S.A., Martin, K.D., McManus, E., Molnar, J.,
Recchia, C.A. & Robertson, J. (2007) Marine ecoregions of the world: a
bioregionalization of coastal and shelf areas. BioScience, 57, 573–583.
151
Springer, V.G. (1988) The Indo-Pacific blenniid fish genus Ecsenius. Smithsonian
Contributions to Zoology, 465, 1–134.
Springer, V.G. & Williams, J.T. (1994) The Indo-West Pacific blenniid fish genus Istiblennius
reappraised: a revision of Istiblennius, Blenniella, and Paralticus, new genus.
Smithsonian Contributions to Zoology, 565, 1–193.
Steininger, F.F. & Rögl, F. (1984) Paleogeography and palinspastic reconstruction of the
Neogene of the Mediterranean and Paratethys. The Geological Evolution of the Eastern
Mediterranean (ed. by J.E. Dixon and A.H. Robertson), pp. 659–668. Blackwell,
Oxford.
Stobutzki, I.C. & Bellwood, D.R. (1997) Sustained swimming abilities of the late pelagic stages
of coral reef fishes. Marine Ecology Progress Series, 149, 35–41.
Strathmann, R.R., Hughes, T.P., Kuris, A.M., Lindeman, K.C., Morgan, S.G., Pandolfi, J.M. &
Warner, R.R. (2002) Evolution of local recruitment and its consequences for marine
populations. Bulletin of Marine Science, 70, 377–396.
Streelman, J.T. & Karl, S.A. (1997) Reconstructing labroid evolution with single–copy nuclear
DNA. Proceedings of the Royal Society B: Biological Sciences, 264, 1011–1020.
Streelman, J.T., Alfaro, M., Westneat, M.W., Bellwood, D.R. & Karl, S.A. (2002) The
evolutionary history of the parrotfishes: biogeography, ecomorphology, and
comparative diversity. Evolution, 56, 961–971.
Sugiura, N. (1978) Further analysis of the data by Akaike' s information criterion and the finite
corrections. Communications in Statistics – Theory and Methods, 7, 13–26.
Swearer, S.E., Shima, J.S., Hellberg, M.E., Thorrold, S.R., Jones, G.P., Robertson, D.R.,
Morgan, S.G., Selkoe, K.A., Ruiz, G.M. & Warner, R.R. (2002) Evidence of self-
recruitment in demersal marine populations. Bulletin of Marine Science, 70, 251–271.
Swofford, D.L. (2003) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods).
Version 4. Sinauer Associates, Sunderland.
Taylor, C.M. & Gotelli, N.J. (1994) The macroecology of Cyprinella: correlates of phylogeny,
body size, and geographical range. The American Naturalist, 144, 549–569.
152
Teske, P.R., Hamilton, H., Matthee, C.A. & Barker, N.P. (2007) Signatures of seaway closures
and founder dispersal in the phylogeny of a circumglobally distributed seahorse lineage.
BMC Evolutionary Biology, 7, 138.
Thresher, R.E. & Brothers, E.B. (1985) Reproductive ecology and biogeography of Indo-West
Pacific angelfishes (Pisces: Pomacanthidae). Evolution, 39, 878–887.
Toussaint, E.F.A., Sagata, K., Surbakti, S., Hendrich, L. & Balke, M. (2013) Australasian sky
islands act as a diversity pump facilitating peripheral speciation and complex reversal
from narrow endemic to widespread ecological supertramp. Ecology and Evolution, 3,
1031–1049.
van Herwerden, L., Choat, J.H., Dudgeon, C.L., Carlos, G., Newman, S.J., Frisch, A. & van
Oppen, M. (2006) Contrasting patterns of genetic structure in two species of the coral
trout Plectropomus (Serranidae) from east and west Australia: introgressive
hybridisation or ancestral polymorphisms. Molecular Phylogenetics and Evolution, 41,
420–435.
Vari, R.P. (1978) The terapon perches (Percoidei, Teraponidae). A cladistic analysis and
taxonomic revision. Bulletin of the American Museum of Natural History, 159, 175–
340.
Veron, J.E.N., Devantier, L.M., Turak, E., Green, A.L., Kininmonth, S., Stafford-Smith, M. &
Peterson, N. (2009) Delineating the Coral Triangle. Galaxea, Journal of Coral Reef
Studies, 11, 91–100.
Vivekanandan, E., Hermes, R. & O'Brien, C. (2012) Climate change effects in the Bay of
Bengal large marine ecosystem. Frontline Observations on Climate Change and
Sustainability of Large Marine Ecosystems (ed. by K. Sherman and G. McGovern), pp.
97–111. U.N.D. Programme, New York.
Wagner, W.L. & Funk, V.A. (1995) Hawaiian Biogeography. Evolution on a Hot Spot
Archipelago. Smithsonian Institution Press, Washington.
153
Wainwright, P.C., Bellwood, D.R., Westneat, M.W., Grubich, J.R. & Hoey, A.S. (2004) A
functional morphospace for the skull of labrid fishes: patterns of diversity in a complex
biomechanical system. Biological Journal of the Linnean Society, 82, 1–25.
Wallace, A.R. (1889) Darwinism: An Exposition of the Theory of Natural Selection with Some
of Its Applications, 2d edn. Macmillan, London.
Wallace, C.C., Paulay, G., Hoeksema, B.W., Bellwood, D.R., Hutchings, P.A., Barber, P.H.,
Erdmann, M. & Wolstenholme, J. (2000) Nature and origins of unique high diversity
reef faunas in the Bay of Tomini, Central Sulawesi: the ultimate "centre of diversity"?
Proceedings of the 9th International Coral Reef Symposium, Bali, 1, 185–192.
Webb, T.J. & Gaston, K.J. (2000) Geographic range size and evolutionary age in birds.
Proceedings of the Royal Society B: Biological Sciences, 267, 1843–1850.
Weber, M.d.M., Stevens, R.D., Lorini, M.L. & Grelle, C.E.V. (2014) Have old species reached
most environmentally suitable areas? A case study with South American phyllostomid
bats. Global Ecology and Biogeography.
Wellman, P. (1983) Hotspot volcanism in Australia and New Zealand: Cainozoic and mid-
Mesozoic. Tectonophysics, 96, 225–243.
Westneat, M.W. & Alfaro, M.E. (2005) Phylogenetic relationships and evolutionary history of
the reef fish family Labridae. Molecular Phylogenetics and Evolution, 36, 370–390.
Wilgenbusch, J.C., Warren, D.L. & Swofford, D.L. (2004) AWTY: A System for Graphical
Exploration of MCMC Convergence in Bayesian Inference. Available at:
http://ceb.csit.fsu.edu/awty/.
Williams, S.T. & Reid, D.G. (2004) Speciation and diversity on tropical rocky shores: a global
phylogeny of snails of the genus Echinolittorina. Evolution, 58, 2227–2251.
Willis, J.C. (1922) Age and Area: A Study in Geographical Distribution and Origin of Species.
Cambridge University Press, London.
Wilson, E.O. (1961) The nature of the taxon cycle in the Melanesian ant fauna. The American
Naturalist, 95, 169–193.
154
Wilson, J.T. (1963) A possible origin of the Hawaiian Islands. Canadian Journal of Physics,
41, 863–870.
Winterbottom, R. (1986) Revision and vicariance biogeography of the subfamily
Congrogadinae (Pisces: Perciformes: Pseudochromidae). Indo-Pacific Fishes, 9, 1–34.
Winterbottom, R., Reist, J.D. & Goodchild, C.D. (1984) Geographic variation in Congrogadus
subducens (Teleostei, Perciformes, Congrogadidae). Canadian Journal of Zoology, 62,
1605–1617.
Winters, K.L., van Herwerden, L., Choat, J.H. & Robertson, D.R. (2010) Phylogeography of the
Indo-Pacific parrotfish Scarus psittacus: isolation generates distinctive peripheral
populations in two oceans. Marine Biology, 157, 1679–1691.
Wollenberg, K.C., Vieites, D.R., Glaw, F. & Vences, M. (2011) Speciation in little: the role of
range and body size in the diversification of Malagasy mantellid frogs. BMC
Evolutionary Biology, 11, 217.
Woodland, D.J. (1983) Zoogeography of the Siganidae (Pisces): an interpretation of distribution
and richness patterns. Bulletin of Marine Science, 33, 713–717.
Woodland, D.J. (1986) Wallace's Line and the distribution of marine inshore fishes. Indo-
Pacific Fish Biology: Proceedings of the 2nd International Conference on Indo-Pacific
Fishes (ed. by T. Uyeno, R. Arai, T. Taniuchi and K. Matsuura), pp. 453–460.
Ichthyological Society of Japan, Tokyo.
Woodroffe, C.D., Brooke, B.P., Linklater, M., Kennedy, D.M., Jones, B.G., Buchanan, C.,
Mleczko, R., Hua, Q. & Zhao, J.-x. (2010) Response of coral reefs to climate change:
expansion and demise of the southernmost Pacific coral reef. Geophysical Research
Letters, 37, L15602.
Yaakub, S.M., Bellwood, D.R., van Herwerden, L. & Walsh, F.M. (2006) Hybridization in
coral reef fishes: introgression and bi-directional gene exchange in Thalassoma (family
Labridae). Molecular Phylogenetics and Evolution, 40, 84–100.
155
Zhisheng, A., Kutzbach, J.E., Prell, W.L. & Porter, S.C. (2001) Evolution of Asian monsoons
and phased uplift of the Himalaya-Tibetan plateau since Late Miocene times. Nature,
411, 62–66.
Zuckerkandl, E. & Pauling, L. (1965) Evolutionary divergence and convergence in proteins.
Evolving Genes and Proteins (ed. by V. Bryson & H. Vogel), pp. 97–166. Academic
Press, New York.
Zwickl, D.J. (2006) Genetic algorithm approaches for the phylogenetic analysis of large
biological sequence datasets under maximum likelihood criterion. Ph.D. Thesis, The
University of Texas, Austin.
156
Appendix A: Supplementary materials for Chapter 2 The following supplemental accompanies general methods for the articles
The role of peripheral endemism in species diversification: evidence from the coral reef fish genus Anampses (Family: Labridae)
Jennifer R. Hodge 1,2,3, Charmaine I. Read 1,3, Lynne van Herwerden 1,3, David R. Bellwood 1,2
1 School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2 Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia 3 Molecular Evolution and Ecology Lab, James Cook University, Townsville, Qld, 4811, Australia
Molecular Phylogenetics and Evolution 62(2): 653–663 (2012)
Evolution of sympatric species: a case study of the coral reef fish genus Pomacanthus (Pomacanthidae)
Jennifer R. Hodge 1,2,3, Charmaine I. Read 1,3, David R. Bellwood 1,2, Lynne van Herwerden 1,3
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia 3Molecular Evolution and Ecology Lab, James Cook University, Townsville, Qld, 4811, Australia
Journal of Biogeography 40(9): 1676–1687 (2013)
157
On the relationship between species age and geographical range in reef fishes: are widespread species older than they seem?
Jennifer R. Hodge 1,2, David R. Bellwood 1,2
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia
Global Ecology and Biogeography (2014 in press)
Temporal evolution of coral reef fishes: global patterns and disparity in isolated locations
Jennifer R. Hodge 1,2,3, Lynne van Herwerden 1,3, David R. Bellwood 1,2
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia 3Molecular Evolution and Ecology Lab, James Cook University, Townsville, Qld, 4811, Australia
Journal of Biogeography 41(11): 2115–2127 (2014)
The geography of speciation in coral reef fishes
Jennifer R. Hodge 1,2, David R. Bellwood 1,2
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia
Submitted for publication
158
TABLES
Table S2.1 Locality information and GenBank accession numbers for specimens used in the
Anampses case study (Chapter 3). Includes representative sister group and outgroup species.
Table S2.2 Collection localities and GenBank accession numbers for specimens used in the
Pomacanthus case study (Chapter 4). Includes representative sister group and outgroup species.
Table S2.3 GenBank accession numbers for species included in the multi-family phylogenetic
analyses (Chapters 5–7).
Table S2.4 Details of fossil data used for temporal-calibration of the multi-family phylogenetic
analyses (Chapters 5–7) in BEAST.
Table S2.5 Partition scheme and models of evolution selected by PARTITIONFINDER under the
Bayesian Information Criterion (BIC) used in the multi-family phylogenetic analyses (Chapters
5–7).
REFERENCE LIST
159
Table S2.1 Locality information and GenBank accession numbers for specimens used in the Anampses case study (Chapter 3). Includes representative sister group and
outgroup species. (—) = location/sequence unavailable. Where possible individual specimens were sampled from different locations.
Specimen ID Genus Species Collection Locations GenBank Accession Numbers Ingroup taxa 12s CO1 S7 109 Anampses caeruleopunctatus Moorea JN935296 JN935324 JN935348 822 Lizard Island, GBR JN935297 JN935325 JN935349 1431 Mauritius JN935298 JN935326 JN935350 940 Anampses chrysocephalus Hawaii JN935299 JN935327 JN935351 1293 Hawaii JN935300 — — 05 Anampses cuvier Hawaii JN935301 JN935328 JN935352 06 Hawaii JN935302 JN935329 JN935353 956 Anampses elegans Lord Howe Island, NSW JN935303 JN935330 JN935354 974 Lord Howe Island, NSW JN935304 JN935331 JN935355 979 Anampses femininus Lord Howe Island, NSW JN935305 JN935332 JN935356 776 Anampses geographicus Day Reef, GBR JN935306 JN935333 JN935357 825 Lizard Island, GBR JN935307 JN935334 JN935358 1282 Anampses lennardi Gove Peninsula, NT JN935308 JN935335 JN935359 1291 Anampses lineatus Bali JN935309 JN935336 JN935360 1399 Mauritius JN935310 JN935337 JN935361 1490 Anampses melanurus Vanuatu JN935311 JN935338 JN935362 186 Anampses meleagrides Bird Island, Seychelles JN935312 — JN935363 692 Pohnpei, Micronesia — JN935339 — 1186 Cocos (Keeling) Islands JN935313 JN935340 JN935364 252 Anampses neoguinaicus Yonge Reef, GBR JN935314 JN935341 JN935365 251 Yonge Reef, GBR JN935315 — JN935366 82 Anampses twistii Moorea JN935316 JN935342 JN935367 1170 Cocos (Keeling) Islands JN935317 JN935343 JN935368 Representative sister group taxa 502 a Halichoeres hortulanus Indonesia — FJ583502 — 208 b Indonesia — — EF488031 380 c Indonesia AY850823 AY850761 — 761 d — AY279601 — — 75 Halichoeres marginatus Trunk Reef, GBR JN935318 JN935344 JN935369 230 Pelorus Island, GBR JN935319 — —
160 Specimen ID Genus Species Collection Locations GenBank Accession Numbers Representative sister group taxa cont. 12s CO1 S7 511 a — — FJ583511 — 1138 Macropharyngodon bipartitus Maldives JN935320 JN935345 EF488039 1139 Maldives JN935321 JN935346 EF488040 1373 Macropharyngodon cyanoguttatus Mauritius JN935322 — EF488041 1374 Mauritius JN935323 JN935347 EF488042 Outgroup taxa 38 e Symphodus melops Portugal AY092049 — — 197 e Portugal AF414197 — — 607 f Scotland — — — 621 g — — GQ341604 GQ341621
Publication source for sequences obtained from GenBank: a Steinke et al. (2009); b Yaakub et al. (2007); c Barber and Bellwood (2005); d Westneat and Alfaro (2005); e Almada et al. (2002); f Hanel et al. (2002); g Kazancıoğlu et al. (2009).
161
Table S2.2 Collection localities and GenBank accession numbers for specimens used in the Pomacanthus case study (Chapter 4). Includes representative sister group and
outgroup species. Analyses were preformed on consensus sequences; where individual specimens were sampled from different locations, each location is listed. (—) =
location/sequence unavailable. GBR, Great Barrier Reef.
GenBank accession numbers Genus Species Collection locations 12S 16S S7 Ingroup taxa Pomacanthus annularis Sri Lanka/Indonesia KC820895 KC820905 KC820915 Pomacanthus arcuatus Caribbean KC820896 KC820906 KC820916 Pomacanthus asfur Red Sea AY530826 a AY530854 a KC820917 Pomacanthus chrysurus — KC820897 KC820907 KC820918 Pomacanthus imperator Sri Lanka KC820898 KC820908 KC820919 Pomacanthus maculosus Red Sea KC820899 KC820909 KC820920 Pomacanthus navarchus Indonesia KC820900 KC820910 KC820921 Pomacanthus paru Belize AY530824 a AY530852 a KC820922 Pomacanthus rhomboides South Africa KC820901 KC820911 KC820923 Pomacanthus semicirculatus Indonesia/One Tree Island, GBR AF108574 b AY530844 a KC820924 Pomacanthus sexstriatus Lizard Island, GBR AF108575 b AY530858 a KC820925 Pomacanthus xanthometopon Philippines/Indonesia KC820902 KC820912 KC820926 Pomacanthus zonipectus Sea of Cortez AY530840 a AY530874 a KC820927 Representative sister group taxa Pygoplites diacanthus Moorea/Younge Reef, GBR AF108577 b AY530873 a KC820928 Holacanthus africanus Cape Verde Islands/São Tomé KC845393 c KC845330 c KC845372 c Holacanthus tricolor Ceará State, Brazil/Marathon Key, Florida KC845399 c KC845336 c KC845378 c Holacanthus limbaughi Clipperton Atoll KC845391 c KC845328 c KC845370 c Chaetodontoplus duboulayi Mackay, GBR KC820903 KC820913 KC820929 Chaetodontoplus mesoleucus Philippines/Indonesia KC820904 KC820914 KC820930 Outgroup taxa Acanthurus nigricans Hick’s Reef, GBR/Cocos (Keeling) Island AY057239 d AY057286 d EF648230 e Naso lituratus —/West Australia AF055603 f AF055624 f EF648256 e Zanclus cornutus — EF616898 g EF616980 g —
162 GenBank accession numbers
Genus Species Collection locations 12S 16S S7 Outgroup taxa cont. Luvarus imperialis —/Australian Museum, Sydney AF055601 f AY264587 h —
Platax orbicularis Philippines AF055597 f/AY279562 h AF055618 f/AY279665 i FJ167848 j
Publication source for sequences obtained from GenBank: a Bellwood et al. (2004); b Nelson et al. (unpublished), location unavailable; c Alva-Campbell et al. (2010); d Clements et al. (2003); e Marie et al. (2007); f Tang et al. (1999), location unavailable; g Fessler & Westneat (2007); h Klanten et al. (2004); i Westneat & Alfaro (2005), location unavailable; j Bellwood et al. (2010).
163
Table S2.3 GenBank accession numbers for species included in the multi-family phylogenetic analyses (Chapters 5–7). (—) = unavailable sequence. * sequences courtesy of
K. Ma.
Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Chaetodontidae Amphichaetodon howensis AY530860 AY530830 — FJ167682 — FJ167793 Amphichaetodon melbae EF616904 EF616820 — — EF617151 — Chaetodon adiergastos EF616942 EF616859 — — EF617190 — Chaetodon argentatus — AF108505 FJ583020 U23581 — FJ167794 Chaetodon aureofasciatus — AF108506 — FJ167683 — FJ167795 Chaetodon auriga EF616943 EF616860 JF434764 JF457933 EF617191 — Chaetodon auripes AP006004 AP006004 FJ583027 FJ167684 — FJ167796 Chaetodon austriacus EF616910 EF616826 — AF108583 EF617157 — Chaetodon baronessa EF616957 EF616874 — FJ167685 EF617205 FJ167797 Chaetodon bennetti EF616911 EF616827 JF434768 JF457936 EF617158 FJ167798 Chaetodon blackburnii JF457269 — JF434770 JF457938 — FJ167799 Chaetodon burgessi EF616934 EF616851 FJ583031 — EF617182 — Chaetodon capistratus EF616914 EF616830 JQ839993 FJ167688 EF617161 FJ167800 Chaetodon citrinellus EF616950 EF616867 JF434772 JF457940 EF617198 FJ167801 Chaetodon collare EF616924 EF616840 FJ583043 FJ167689 EF617171 FJ167802 Chaetodon decussatus — AF108512 FJ583045 AF108587 — — Chaetodon dolosus EF616938 EF616855 JF493091 — EF617186 — Chaetodon ephippium EF616926 EF616842 JF434773 FJ167690 EF617173 FJ167803 Chaetodon falcula EF616919 EF616835 JF434776 JF457943 EF617166 — Chaetodon fasciatus EF616940 EF616857 — — EF617188 — Chaetodon flavirostris — AF108514 FJ583053 AF108589 — — Chaetodon fremblii EF616925 EF616841 — FJ973160 EF617172 FJ167804 Chaetodon guentheri EF616951 EF616868 — — EF617199 — Chaetodon guttatissimus JF457281 AF108515 JF434782 JF457949 — — Chaetodon hoefleri EF616908 EF616824 — — EF617155 — Chaetodon humeralis HM778172 — — FJ167692 — FJ167805 Chaetodon interruptus — — — FJ167693 — FJ167806 Chaetodon kleinii EF616909 EF616825 JF434785 JF457952 EF617156 FJ167807 Chaetodon larvatus EF616930 EF616847 — AF108592 EF617178 —
164 Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Chaetodontidae cont. Chaetodon lineolatus EF616939 EF616856 — AF108593 EF617187 FJ167808 Chaetodon lunula EF616944 EF616861 JF434792 JF457959 EF617192 FJ167809 Chaetodon lunulatus JF457293 AJ748307 JF434794 JF457961 — FJ167810 Chaetodon madagaskariensis — AJ748308 JF493098 U23598 — — Chaetodon marleyi — — JF493101 — — — Chaetodon melannotus EF616945 EF616862 JF434800 JF457967 EF617193 FJ167811 Chaetodon mertensii EF616958 EF616875 JF43480 JF457969 EF617206 — Chaetodon mesoleucos — AF108521 — AF108596 — — Chaetodon meyeri EF616936 EF616853 JF434804 JF457971 EF617184 FJ167812 Chaetodon miliaris EF616918 EF616834 — FJ973249 EF617165 FJ167813 Chaetodon mitratus — — JF493105 — — — Chaetodon multicinctus EF616922 EF616838 FJ583067 FJ973301 EF617169 FJ167814 Chaetodon ocellatus EF616933 EF616850 JQ839994 FJ167698 EF617181 FJ167815 Chaetodon ocellicaudus EF616946 EF616863 FJ583069 AF108598 EF617194 — Chaetodon octofasciatus EF616952 EF616869 — AF108599 EF617200 — Chaetodon ornatissimus EF616953 EF616870 JF434807 JF457974 EF617201 FJ167816 Chaetodon oxycephalus EF616929 EF616845 — AF108601 EF617176 — Chaetodon paucifasciatus EF616937 EF616854 — U23607 EF617185 — Chaetodon pelewensis EF616932 EF616849 JF434809 JF457976 EF617180 — Chaetodon plebeius — EF616846 JQ362391 AF108602 EF617177 FJ167817 Chaetodon punctatofasciatus EF616931 EF616848 FJ583074 U23619 EF617179 — Chaetodon quadrimaculatus EF616923 EF616839 JF434811 JF457978 EF617170 FJ167818 Chaetodon rafflesii EF616947 EF616864 FJ583082 AF108604 EF617195 — Chaetodon rainfordi EF616935 EF616852 — FJ167699 EF617183 FJ167819 Chaetodon reticulatus EF616916 EF616852 JF434813 JF457980 EF617163 FJ167820 Chaetodon robustus EF616921 EF616837 — FJ167701 EF617168 FJ167821 Chaetodon sanctaehelenae — — — FJ167702 — FJ167822 Chaetodon sedentarius EF616912 EF616828 — FJ167703 EF617159 FJ167823 Chaetodon selene EF616948 EF616865 — — EF617196 — Chaetodon semeion EF616913 EF616829 FJ583083 — EF617160 — Chaetodon semilarvatus EF616920 EF616836 — AJ748305 EF617167 — Chaetodon speculum EF616917 EF616833 FJ583084 AF108606 EF617164 — Chaetodon striatus EF616915 EF616831 JQ840450 FJ167704 EF617162 FJ167824
165
Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Chaetodontidae cont. Chaetodon tinkeri — — — U23629 — FJ167825 Chaetodon trichrous JF457314 AJ748312 JF434815 JF457982 — — Chaetodon tricinctus — — — FJ167706 — FJ167826 Chaetodon trifascialis EF616927 EF616843 JF434822 JF457989 EF617174 FJ167827 Chaetodon trifasciatus EF616954 EF616871 JF434830 JF457997 EF617202 FJ167828 Chaetodon ulietensis EF616928 EF616844 JF434832 JF457999 EF617175 FJ167829 Chaetodon unimaculatus EF616955 EF616872 JF434836 JF458003 EF617203 FJ167830 Chaetodon vagabundus EF616949 EF616866 JF434842 JF458009 EF617197 — Chaetodon wiebeli — — FJ583110 — — — Chaetodon xanthocephalus — — JF493107 — — — Chaetodon xanthurus EF616956 EF616873 JF434845 JF458012 EF617204 — Chaetodon zanzibarensis EF616941 EF616858 JF434846 JF458013 EF617189 — Chelmon marginalis EF616960 EF616877 — FJ167711 EF617208 FJ167831 Chelmon muelleri — — — FJ167712 — FJ167832 Chelmon rostratus EF616959 EF616876 FJ583134 AF108612 EF617207 FJ167833 Chelmonops curiosus — — — FJ167713 — FJ167834 Chelmonops truncatus EF616961 EF616878 — FJ167714 EF617209 FJ167835 Coradion altivelis — AF108538 — AF108613 — FJ167836 Coradion chrysozonus EF616963 EF616880 — AF108614 EF617211 FJ167837 Coradion melanopus EF616962 EF616879 — — EF617210 — Forcipiger flavissimus EF616964 EF616881 JF434973 JF458132 EF617212 FJ167838 Forcipiger longirostris EF616965 EF616882 JF434975 FJ167715 EF617213 FJ167839 Hemitaurichthys polylepis EF616906 EF616822 JF435020 JF458156 EF617153 FJ167840 Hemitaurichthys thompsoni JF457508 — JF435022 JF458158 — FJ167841 Hemitaurichthys zoster EF616966 EF616883 JF435024 JF458160 EF617214 — Heniochus acuminatus EF616907 EF616823 JF435025 JF458161 EF617154 FJ167842 Heniochus chrysostomus JF457512 EF616884 JF435027 JF458163 — FJ167843 Heniochus diphreutes EF616970 EF616888 JF435031 JF458167 EF617218 — Heniochus monoceros EF616968 EF616886 JF435033 JF458169 EF617216 — Heniochus pleurotaenia — AF108545 — AF108620 — — Heniochus singularius EF616967 EF616885 FJ583547 FJ167718 EF617215 FJ167844 Heniochus varius EF616969 EF616887 FJ583549 AF108621 EF617217 — Johnrandallia nigrirostris EF616971 EF616889 — FJ167719 — FJ167845
166 Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Chaetodontidae cont. Parachaetodon ocellatus — AF108547 — AF108622 — FJ167847 Labridae Anampses caeruleopunctatus KJ866392 JN935296 JN935325 JF457906 AY279781 JN935348 Anampses chrysocephalus KJ866393 JN935299 JN935327 — — JN935351 Anampses cuvier KJ866394 JN935301 JN935328 — — JN935352 Anampses elegans KJ866395 JN935303 JN935330 — — JN935354 Anampses femininus KJ866396 JN935305 JN935332 — — JN935356 Anampses geographicus KJ866397 JN935306 JN935333 — — JN935357 Anampses lennardi KJ866398 JN935308 JN935335 — — JN935359 Anampses lineatus KJ866399 JN935309 JN935336 JF457908 — JN935361 Anampses melanurus KJ866400 JN935311 JN935338 — — JN935362 Anampses meleagrides KJ866401 JN935312 JN935339 — — JN935363 Anampses neoguinaicus KJ866402 JN935314 JN935341 — AY279782 JN935366 Anampses twistii KJ866403 JN935317 JN935343 JF457909 — JN935367 Bolbometopon muricatum JX026453 AY081073 — EU601357 AY081108 JX026592 Centrolabrus caeruleus AY092034 AF414192 — — — GQ341613 Centrolabrus exoletus AY092041 EU601224 GQ341588 EU601403 EU601303 — Centrolabrus trutta AY092035 AF414195 — — — GQ341614 Chlorurus atrilunula JX026457 — — — — JX026596 Chlorurus bleekeri JX026458 EU601182 — EU601361 EU601267 JX026597 Chlorurus bowersi EU601231 EU601183 — EU601362 EU601268 JX026598 Chlorurus capistratoides JX026460 EU601184 — EU601363 EU601269 JX026599 Chlorurus cyanescens JX026461 — — — — JX026600 Chlorurus enneacanthus JX026462 — — — — JX026601 Chlorurus frontalis JX026463 — JQ431620 — — JX026602 Chlorurus gibbus JX026464 — FJ237699 — — — Chlorurus japanensis JX026465 EU601209 — EU601388 EU601288 JX026603 Chlorurus microrhinos JX026466 EU601185 — EU601364 EU601270 JX026604 Chlorurus oedema JX026467 EU601186 — EU601365 AY081107 JX026605 Chlorurus perspicillatus JX026468 — — — — JX026606 Chlorurus rhakoura JX026469 — — — — JX026607 Chlorurus sordidus JX026471 EU601187 JQ349891 EU601366 AY081106 JX026609 Chlorurus spilurus JX026470 — — — — JX026608
167
Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Labridae cont. Chlorurus strongylocephalus JX026472 — — — — JX026610 Cryptotomus roseus EU601235 AY279592 JQ841518 EU601367 EU601271 — Diproctacanthus xanthurus AY279698 AY279595 — — AY279801 — Epibulus brevis EF520654 — — — — — Epibulus insidiator AY279699 AJ810132 JF434969 JF458128 EU601301 — Gomphosus caeruleus JF457465 — JF434979 JF458133 — AY329640 Gomphosus varius JF457467 AY279597 JF434981 AY328858 AY279803 AY329641 Haletta semifasciata AY662708 AY279656 EF609368 — AY279862 — Hemigymnus fasciatus JF457502 AJ810136 JF435016 JF458152 — — Hemigymnus melapterus JF457504 AJ810137 JF435018 JF458154 AY279817 — Hipposcarus harid JX026455 — — — — JX026594 Hipposcarus longiceps JX026456 EU601189 — EU885924 AY081110 JX026595 Labrichthys unilineatus AY850875 AJ810138 JF493715 — AY279819 — Lachnolaimus maximus AY857946 AY279618 JQ839821 EU601404 AY662809 — Larabicus quadrilineatus AY279722 AY279619 — — AY279825 — Leptoscarus vaigiensis AY081094 EU601190 FJ583627 EU601369 AY081111 — Neodax balteatus AY279760 AY279657 — — AY279863 — Odax acroptilus AY279761 AY279658 — — AY279864 — Odax cyanoallix AY279762 AY279659 — — AY279865 — Odax cyanomelas AY279763 AY279660 4850752 4850753 AY279866 — Odax pullus AY279764 AY279661 — — AY279867 — Ophthalmolepis lineolata AY279731 AY279628 JQ839560 — AY279834 — Oxyjulis californica AY279735 AY279632 JN582150 DQ132499 AY279838 — Pseudodax moluccanus JF457633 AY279636 JF435149 JF458248 AY279842 — Scarus altipinnis JX026473 EU601192 JQ432095 EU601371 EU601273 JX026611 Scarus arabicus JX026474 — — — — JX026612 Scarus chameleon JX026475 EU601193 FJ237917 EU601372 EU601274 JX026613 Scarus coelestinus AY081084 EU601194 — EU601373 AY081101 JX026614 Scarus coeruleus JX026476 — — — — JX026615 Scarus collana JX026477 — — — — JX026616 Scarus compressus JX026478 — — — — JX026617 Scarus dimidiatus JX026479 EU601195 — EU601374 EU601275 JX026618 Scarus dubius JX026480 — — — — JX026619
168 Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Labridae cont. Scarus falcipinnis JX026481 — — — — JX026620 Scarus ferrugineus JX026482 — — — — JX026621 Scarus festivus JX026483 EU601196 — EU601375 EU601276 JX026622 Scarus flavipectoralis JX026484 EU601197 — EU601376 AY081103 JX026623 Scarus forsteni JX026485 EU601198 JQ432097 EU601377 EU601278 JX026624 Scarus frenatus JX026486 EU601199 FJ237920 EU601378 AY081104 JX026625 Scarus fuscopurpureus JX026487 — — — — JX026626 Scarus ghobban EU601241 EU601200 EF609452 EU601379 EU601279 JX026629 Scarus globiceps JX026492 EU601201 JQ350330 EU601380 EU601280 JX026631 Scarus guacamaia EU601243 EU601202 JQ843039 EU601381 AY081102 JX026632 Scarus hoefleri JX026493 AY141393 — — — JX026633 Scarus hypselopterus EU601245 EU601204 — EU601383 EU601283 — Scarus iseri JX026494 EU601203 JQ840986 EU601382 EU601282 JX026634 Scarus koputea JX026495 — — — — JX026635 Scarus longipinnis JX026496 — — — — JX026636 Scarus niger JX026497 EU601205 JQ350332 EU601384 EU601284 JX026637 Scarus oviceps JX026498 EU601206 JQ432107 EU601385 EU601285 JX026638 Scarus ovifrons JX026499 — — — — JX026639 Scarus perrico JX026500 — — — — JX026640 Scarus persicus JX026501 — — — — JX026641 Scarus prasiognathos JX026502 EU601207 — EU601386 EU601286 JX026642 Scarus psittacus JX026503 EU601208 JQ432110 EU601387 EU601287 JX026643 Scarus quoyi JX026504 EU601210 — EU601389 EU601289 JX026644 Scarus rivulatus JX026505 EU601211 — EU601390 EU601290 JX026645 Scarus rubroviolaceus EU601253 EU601212 GU805008 EU601391 EU601291 JX026646 Scarus russelii JX026510 — — — — JX026650 Scarus scaber JX026511 — JQ350334 — — JX026651 Scarus schlegeli JX026512 EU601213 JQ432114 EU601392 EU601292 JX026652 Scarus spinus JX026513 EU601214 — EU601393 EU601293 JX026653 Scarus taeniopterus JX026514 EU601215 JQ842302 EU601394 EU601294 JX026654 Scarus tricolor JX026515 EU601216 JQ350335 EU601395 EU601295 JX026655 Scarus trispinosus JX026516 — — — — JX026656 Scarus vetula JX026517 — FJ584084 — — JX026657
169
Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Labridae cont. Scarus viridifucatus JX026518 — — — — JX026658 Scarus xanthopleura JX026519 — — — — JX026659 Scarus zelindae JX026520 — — — — JX026660 Scarus zufar JX026521 — — — — JX026661 Symphodus bailloni AY092037 AY092052 GQ341601 — — GQ341616 Symphodus cinereus AY092036 AJ810147 GQ341603 — — GQ341618 Symphodus doderleini AF517602 — GQ341606 — — — Symphodus mediterraneus AF517601 AJ810148 GQ341607 — — GQ341620 Symphodus melanocercus AF517595 AJ810149 GQ341600 — — — Symphodus melops AY092038 AF414197 GQ341604 — — GQ341621 Symphodus ocellatus AF517603 AJ810150 — — — — Symphodus roissali AY092039 AJ810151 GQ341605 — — GQ341622 Symphodus rostratus AY092040 AF414198 GQ341608 — — GQ341617 Symphodus tinca AF517596 AJ810152 GQ341602 — — GQ341619 Tautoga onitis AF517588 AY279648 GQ341609 EU601402 AY662810 — Tautogolabrus adspersus AF517585 AY279649 GQ341610 HM049968 AY279855 — Thalassoma amblycephalum JF457667 — JF435183 AY328860 — AY329643 Thalassoma ascensionis AY328988 — — AY328861 — AY329644 Thalassoma ballieui AY328989 — DQ521017 AY328862 — AY329645 Thalassoma bifasciatum AY279753 AJ810153 JQ839917 AY328863 AY279856 AY329646 Thalassoma cupido AY328991 — — AY328864 — AY329647 Thalassoma duperrey AY328994 — — AY328865 — DQ443832 Thalassoma genivittatum JF457672 — JF435188 AY328866 — DQ443833 Thalassoma grammaticum AY328993 — JQ839619 AY328867 — AY329650 Thalassoma hardwicke AY850865 AY850802 JF435196 AY328868 — DQ443821 Thalassoma hebraicum JF457683 — JF435199 AY328869 — AY329652 Thalassoma jansenii AY328997 — FJ459566 AY328870 — DQ443820 Thalassoma loxum AY328998 — — AY328871 — AY329654 Thalassoma lucasanum AY328999 — JQ839621 AY328872 — AY329655 Thalassoma lunare JQ178236 AJ810154 JF435203 AY328873 AY279857 DQ443836 Thalassoma lutescens AY850863 AJ810155 FJ584187 AY328874 — DQ443835 Thalassoma newtoni AY329002 — — AY328875 — AY329658 Thalassoma noronhanum AY329003 — JQ839633 AY328876 — AY329659
170 Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Labridae cont. Thalassoma pavo AF517583 — GQ341611 DQ198011 — AY329660 Thalassoma purpureum JF457693 — JF435209 AY328878 — AY329661 Thalassoma quinquevittatum AY850864 AY850801 JF435210 AY328879 — DQ443802 Thalassoma robertsoni AY329007 — JQ839646 AY328880 — AY329663 Thalassoma rueppellii AY329008 — — AY328881 — AY329664 Thalassoma sanctaehelenae AY329009 — — AY328882 — AY329665 Thalassoma septemfasciatum AY329010 — — AY328883 — AY329666 Thalassoma trilobatum JF457696 — JF435212 AY328884 — AY329667 Thalassoma virens AY329012 — JQ839654 AY328885 — AY329668 Xiphocheilus typus AY279756 AY279653 — — AY279859 — Pomacanthidae Holacanthus africanus KC845330 KC845393 — KC845351 — KC845372 Holacanthus bermudensis KC845337 KC845400 — KC845358 EF617231 KC845379 Holacanthus ciliaris KC845332 KC845395 JQ841232 KC845353 — KC845385 Holacanthus clarionensis KC845327 KC845390 — KC845348 — KC845369 Holacanthus limbaughi KC845328 KC845391 — KC845349 — KC845370 Holacanthus passer KC845329 KC845392 — KC845350 — KC845371 Holacanthus tricolor KC845336 KC845399 JQ840536 KC845357 — KC845384 Pomacanthus annularis KC820905 KC820895 FJ583875 AF108646 — KC820915 Pomacanthus arcuatus KC820906 KC820896 JQ841953 — EF617229 KC820916 Pomacanthus asfur AY530854 AY530826 — — — KC820917 Pomacanthus chrysurus KC820907 KC820897 — — — KC820918 Pomacanthus imperator KC820908 KC820898 JQ350235 FJ424070 — KC820919 Pomacanthus maculosus KC820909 KC820899 HQ149906 JN604372 — KC820920 Pomacanthus navarchus KC820910 KC820900 FJ583883 AF108648 — KC820921 Pomacanthus paru AY530852 AY530824 JQ840654 — — KC820922 Pomacanthus rhomboides KC820911 KC820901 HQ945884 FJ167723 — KC820923 Pomacanthus semicirculatus AY530844 AF108574 JQ350237 AF108649 — KC820924 Pomacanthus sexstriatus AY530858 AF108575 — AF108650 — KC820925 Pomacanthus xanthometapon KC820912 KC820902 — AF108651 — KC820926 Pomacanthus zonipectus AY530874 AY530840 GU440472 — — KC820927 Pygoplites diacanthus AY530873 AF108577 JQ350298 AF108652 EF617232 KC820928 Epinephelidae
171
Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Epinephelidae cont. Aethaloperca rogaa AY947565 AY949367 JX674944 EF213721 AY949225 — Alphestes afer AY314003 AY313982 JQ840759 AY313996 AY313992 — Alphestes immaculatus AF297290 AY313980 — AY314002 AY313994 — Alphestes multiguttatus AF297305 AY313981 — AY313995 AY313991 — Anyperodon leucogrammicus AF297306 AY949379 DQ107922 AY963557 EF517738 — Dermatolepis dermatolepis AF297317 AY313984 HM032025 AY314000 AY313988 — Dermatolepis inermis AY314005 AY313979 EU752075 AY314001 AY313987 — Dermatolepis striolata AY314004 AY313983 — AY313999 AY313989 — Gracila albomarginata AY947582 AY949348 — — AY949250 — Mycteroperca acutirostris AY947591 AY949411 GU702334 — AY949251 — Mycteroperca bonaci DQ267147 AY949449 JQ840175 — AY949270 — Mycteroperca fusca AY947597 AY949448 — DQ197968 AY949252 — Mycteroperca interstitialis AY947632 AY949359 FJ583668 — AY949221 — Mycteroperca jordani AF297329 AY949435 GU440412 — AY949303 — Mycteroperca microlepis AF297312 AY949373 JQ842598 — AY949253 — Mycteroperca olfax AF317512 AY949360 — — AY949276 — Mycteroperca phenax AF297303 AY949450 — — AY949265 — Mycteroperca prionura AY947583 AY949361 — — AY949254 — Mycteroperca rosacea AF297300 AY949350 — — AY949268 — Mycteroperca rubra AY947587 AY949364 — DQ197969 AY949255 — Mycteroperca tigris AY947574 AY949452 JQ839849 — AY949217 — Mycteroperca venenosa AF297291 AY949419 JQ839850 — AY949273 — Mycteroperca xenarcha AY947637 AY949445 GU440413 — — — Niphon spinosus HQ731416 AY949420 EF143386 AB108493 AY949210 DQ864743 Paranthias colonus AF297301 AY949351 GU440449 HM049960 — — Paranthias furcifer AY947584 AY949372 JQ365486 — AY949263 — Plectropomus areolatus EF213706 AY949447 JN242595 EF213725 EF517750 * Plectropomus laevis DQ067320 AY949444 JX675027 AY963554 EF517749 * Plectropomus leopardus DQ067321 AY949352 JF750763 AY963555 EF517747 — Plectropomus maculatus AF297320 AY949423 JF750764 EF503636 EF517751 — Plectropomus oligacanthus AY947615 EU930863 * EF213726 EF517748 * Plectropomus pessuliferus — — — AY963553 * * Plectropomus punctatus * * * — — —
172 Mitochondrial loci Nuclear loci Genus Species 16S 12S CO1 CYTB TMO-4C4 S7 Epinephelidae cont. Saloptia powelli AY947631 AY949375 JQ432090 — — KM Variola albimarginata EF213708 AY949412 JX675035 EF213728 AY949261 — Variola louti DQ067319 AY949412 JQ432218 AY786428 EF517745 — Fossil Calibration Taxa Acanrthurus nigricans AY057286 AY057239 EF648266 AY264634 — EF648230 Ambloplites rupestris AY742515 — EU524414 AY225663 — — Archoplites interruptus AY742516 — HQ557524 AY225665 — — Bodianus mesothorax AY279681 AY279578 JQ839406 — AY279784 — Calatomus carolinus AY081092 EU601179 JQ349815 EU601358 AY081109 — Cetoscarus bicolor JX026454 EU601181 JQ349875 EU601360 AY081105 JX026593 Gazza minuta DQ648428 — EU148511 — — — Leiognathus equulus EU366341 DQ533223 EU381032 EU380961 — — Luvarus imperialis AF055622 AF055601 5618975 AB276966 — — Naso lituratus AF055624 AF055603 HM034244 AB276964 — EF648255 Prognathodes aculeatus EF616972 EF616890 — FJ167725 EF617219 FJ167852 Sparisoma chrysopterum AY279748 AY279645 JQ841012 DQ457033 AY081100 — Zanclus cornutus AF055623 AY057235 JQ350417 AB276965 — — Outgroup Taxa Opsanus pardus DQ874745 AF165327 — AF165347 DQ874850 — Porichthys notatus EF119249 AF165333 JQ354294 AF165352 — —
173
Table S2.4 Details of fossil data used for temporal-calibration of the multi-family phylogenetic analyses (Chapters 5–7) in BEAST. Calibrations were placed on nodes
representing the most recent common ancestor (MRCA) of specified genera. H indicates a hard lower bound on the prior distribution; S indicates a soft 95% upper bound.
MRCA Minimum fossil age (Ma) Parametric prior distribution Prior bounds Fossil evidence Archoplites/Ambloplites 15.5 Lognormal 15.5 H – 17.8 S Archoplites clarki 1,2 Bolbometopon/Cetoscarus 5 Lognormal 5 H – 36 S Bolbometopon sp. 3 Calotomus/Sparisoma 14 Lognormal 14 H – 50 S Calotomus preisli 3 Chaetodon/Prognathodes 7.1 Lognormal 7.1 H – 25.2 S Chaetodon ficheuri 4 Gazza/Leiognathus 11.6 Lognormal 11.6 H – 30.1 S Euleiognathus tottori 5,6 Hypsigenyines 50 Exponential 50 H – 95 S Phyllopharyngodon longipinnis 7 Labridae (–Hypsigenyines) 50 Lognormal 50 H – 100 S Eocoris bloti 8, Bellwoodilabrus landinii 9 Luvarus, Zanclus, and Acanthuridae (Acanthurus and Naso) 55.8 Lognormal 55.8 H – 63.9 S Avitoluvarus dianae, Avitoluvarus mariannae,
Kushlukia permira, and Luvarus necopinatus 10 Pseudodax/Bodianus 14 Lognormal 14 H – 50 S Trigondon jugleri 11
Publications sources of fossil evidence: 1 (Smith & Elder, 1985); 2 (Smith & Miller, 1985); 3 (Bellwood & Schultz, 1991); 4 (Carnevale, 2006); 5 (Yabumoto & Uyeno, 1994); 6 (Yabumoto &
Uyeno, 2011); 7 (Bellwood, 1990); 8 (Bannikov & Sorbini, 1990); 9 (Bannikov & Carnevale, 2010); 10 (Bannikov & Tyler, 1995); 11 (Schultz & Bellwood, 2004).
174
Table S2.5 Partition scheme and models of evolution selected by PARTITIONFINDER under the Bayesian
Information Criterion (BIC). The data set consisted of molecular sequences for 312 Percomorph fishes
found throughout the tropical Atlantic and Indo-Pacific; the data were used in the multi-family
phylogenetic analyses (Chapters 5–7). Γ represents the gamma shape parameter.
Data set Base pairs Partition BIC model 16S 569 12S 308 CO1 477 CYTB 92
1 TVM+I+Γ
TMO-4C4 434 2 TrN+I+Γ S7 611 3 TrN+I+Γ
Total 2491 — —
References
Almada, V.C., Almada, F., Henriques, M., Santos, R.S. & Brito, A. (2002) On the phylogenetic
affinities of Centrolabrus trutta and Centrolabrus caeruleus (Perciformes: Labridae) to
the genus Symphodus: molecular, meristic and behavioural evidences. Arquipelago,
19A, 85–92.
Alva-Campbell, Y., Floeter, S.R., Robertson, D.R., Bellwood, D.R. & Bernardi, G. (2010)
Molecular phylogenetics and evolution of Holacanthus angelfishes (Pomacanthidae).
Molecular Phylogenetics and Evolution, 56, 456–461.
Bannikov, A.F. & Carnevale, G. (2010) Bellwoodilabrus landinii n. gen., n. sp., a new genus
and species of labrid fish (Teleostei, Perciformes) from the Eocene of Monte Bolca.
Geodiversitas, 32, 201–220.
Bannikov, A.F. & Sorbini, L. (1990) Eocoris bloti, a new genus and species of labrid fish
(Perciformes, Labroidei) from the Eocene of Monte Bolca, Italy. Studi e Recereche sui
Giacimenti Terziari di Bola, Museo Civico di Storia Naturale, Verona 6, 133–148.
175
Bannikov, A.F. & Tyler, J.C. (1995) Phylogenetic revision of the fish families Luvaridae and
Kushlukiidae (Acanthuroidei), with a new genus and two new species of Eocene
luvarids. Smithsonian Contributions to Paleobiology, 81, 1–45.
Barber, P.H. & Bellwood, D.R. (2005) Biodiversity hotspots: evolutionary origins of
biodiversity in wrasses (Halichoeres : Labridae) in the Indo-Pacific and new world
tropics. Molecular Phylogenetics and Evolution, 35, 235–253.
Bellwood, D.R. (1990) A new fish fossil Phyllopharyngodon longipinnis gen. Et sp. nov.
(Family Labridae) from the Eocene, Monte Bolca, Italy. Studi e Ricerche sui
Giacimenti Terziari di Bolca, Museo Civico di Storia Naturale. Verona, 6, 149–160.
Bellwood, D.R. & Schultz, O. (1991) A review of the fossil record of the parrotfishes
(Labroidei: Scaridae) with a description of a new Calotomus species from the middle
Miocene (Badenian) of Austria. Annalen Naturhistorisches Museum Wien, 92A, 55–71.
Bellwood, D.R., van Herwerden, L. & Konow, N. (2004) Evolution and biogeography of
marine angelfishes (Pisces: Pomacanthidae). Molecular Phylogenetics and Evolution,
33, 140–155.
Bellwood, D.R., Klanten, S., Cowman, P.F., Pratchett, M.S., Konow, N. & van Herwerden, L.
(2010) Evolutionary history of the butterflyfishes (f: Chaetodontidae) and the rise of
coral feeding fishes. Journal of Evolutionary Biology, 23, 335–349.
Carnevale, G. (2006) Morphology and biology of the Miocene butterflyfish Chaetodon ficheuri
(Teleostei: Chaetodontidae). Zoological Journal of the Linnean Society, 146, 251–267.
Clements, K.D., Gray, R.D. & Choat, J.H. (2003) Rapid evolutionary divergences in reef fishes
of the family Acanthuridae (Perciformes: Teleostei). Molecular Phylogenetics and
Evolution, 26, 190–201.
Fessler, J.L. & Westneat, M.W. (2007) Molecular phylogenetics of the butterflyfishes
(Chaetodontidae): taxonomy and biogeography of a global coral reef fish family.
Molecular Phylogenetics and Evolution, 45, 50–68.
176
Hanel R., Westneat, M.W. & Sturmbauer, C. (2002) Phylogenetic relationships, evolution of
broodcare behaviour, and geographic speciation in the wrasse tribe Labrini. Journal of
Molecular Evolution, 55, 776–789.
Kazancıoğlu, E., Near, T.J., Hanel, R. & Wainwright, P.C. (2009) Influence of sexual selection
and feeding functional morphology on diversification rate of parrotfishes (Scaridae).
Proceedings of the Royal Society B: Biological Sciences, 276, 3439–3446.
Klanten, S.O., van Herwerden, L., Choat, J.H. & Blair, D. (2004) Patterns of lineage
diversification in the genus Naso (Acanthuridae). Molecular Phylogenetics and
Evolution, 32, 221–235.
Marie, A.D., van Herwerden, L., Choat, J.H. & Hobbs, J.-P.A. (2007) Hybridization of reef
fishes at the Indo-Pacific biogeographic barrier: a case study. Coral Reefs, 26, 841–850.
Nelson, J.S., Hoddell, R.J., Ch’ng, L.M., Chou, L.M., Kara, A.U., Lim, T.M., Ng, P., Phang,
V.P.E., Chan, W.K. & Lam, T.J. (unpublished direct GenBank submission) Phylogeny
of the Indo-Pacific Chaetodontidae and Pomacanthidae based on mitochondrial
cytochrome b and 12S rDNA sequences.
Schultz, O. & Bellwood, D.R. (2004) Trigonodon oweni and Asima jugleri are different parts of
the same species Trigonodon jugleri, a Chiseltooth Wrasse from the lower and middle
Miocene in Central Europe (Osteichthyes, Labridae, Trigonodontinae). Annalen
Naturhistorisches Museum Wien, 105A, 287–305.
Smith, G.R. & Elder, R.L. (1985) Environmental interpretation of Clarkia fishes. Late Cenozoic
History of the Pacific Northwest (ed. by C.J. Smiley), pp. 85–93. American Association
for the Advancement of Science, San Francisco.
Smith, G.R. & Miller, R.R. (1985) Taxonomy of fishes from Miocene Clarkia Lake beds,
Idaho. Late Cenozoic History of the Pacific Northwest (ed. by C.J. Smiley), pp. 75–83.
American Association for the Advancement of Science, San Francisco, CA.
Steinke, D., Zemlak, T.S. & Hebert, P.D.N. (2009) Barcoding Nemo: DNA-based
identifications for the ornamental fish trade. PLoS One, 4, e6300.
177
Tang, K.L., Berendzen, P.B., Wiley, E.O., Morrissey, J.F., Winterbottom, R. & Johnson, G.D.
(1999) The phylogenetic relationships of the Suborder Acanthuroidei (Teleostei:
Perciformes) based on molecular and morphological evidence. Molecular
Phylogenetics and Evolution, 11, 415–425.
Westneat, M.W. & Alfaro, M.E. (2005) Phylogenetic relationships and evolutionary history of
the reef fish family Labridae. Molecular Phylogenetics and Evolution, 36, 370–390.
Yaakub, S.M., Bellwood, D.R. & van Herwerden, L. (2007) A rare hybridization event in two
common Caribbean wrasses (genus Halichoeres; family Labridae). Coral Reefs, 26,
597–602.
Yabumoto, Y. & Uyeno, T. (1994) A new Miocene ponyfish of the genus Leiognathus (Pisces,
Leiognathidae). Bulletin of the National Science Museum Tokyo Series C, 20, 67–77.
Yabumoto, Y. & Uyeno, T. (2011) Euleiognathus, a new genus proposed for the Miocene
ponyfish, Leiognathus tottori Yabumoto and Uyeno 1994 (Perciformes: Leiognathidae)
from Japan. Ichthyological Research, 58, 19–23.
178
Appendix B: Supplementary materials for Chapter 3 The following supplemental accompanies the article
The role of peripheral endemism in species diversification: evidence from the coral reef fish genus Anampses (Family: Labridae)
Jennifer R. Hodge 1,2,3, Charmaine I. Read 1,3, Lynne van Herwerden 1,3, David R. Bellwood 1,2
1 School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2 Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia 3 Molecular Evolution and Ecology Lab, James Cook University, Townsville, Qld, 4811, Australia
Molecular Phylogenetics and Evolution 62(2): 653–663 (2012)
FIGURES
Figure S3.1 AWTY output from the phylogenetic analysis of Anampses showing pairwise
comparisons of the posterior probabilities of nodes for each of four independent Bayesian
inference analyses implemented in MRBAYES.
Figure S3.2 AWTY output from the phylogenetic analysis of Anampses showing the cumulative
posterior probabilities of the 20 most variable nodes for trees searched by four independent
Markov Chain Monte Carlo analyses implemented in MRBAYES.
Figure S3.3 AWTY output from the phylogenetic analysis of Anampses showing pairwise
comparisons of the posterior probabilities of nodes for each of four independent Bayesian
inference analyses implemented in BEAST.
179
Figure S3.4 AWTY output from the phylogenetic analysis of Anampses showing the cumulative
posterior probabilities of the 15 most variable nodes for trees searched by four independent
Markov Chain Monte Carlo analyses implemented in BEAST.
180 Figure S3.1 AWTY output from the
phylogenetic analysis of Anampses.
Each bivariate plot is a pairwise
comparison of the posterior
probabilities of nodes for each of
four independent Bayesian
inference analyses as implemented
in MRBAYES. Low correlation
among runs would diagnose lack of
convergence, however that was not
detected here.
0
20
40
60
80
100
0 20 40 60 80 100
Run
2
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
3
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
3
Run 2
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 2
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 3
181
Figure S3.2 AWTY output from the
phylogenetic analysis of Anampses.
Each plot shows the cumulative
posterior probabilities of the 20
most variable nodes for trees
searched by four independent
Markov Chain Monte Carlo
analyses as implemented in
MRBAYES; (a) analysis 1; (b)
analysis 2; (c) analysis 3; and (d)
analysis 4. A trend in the frequency
of posterior probabilities for nodes
diagnoses lack of convergence, but
was not observed here.
0
20
40
60
80
100
0 5 10 15 20 25 30 35 40 45 50
Pos
terio
r pro
babi
litie
s of
20
node
s
Number of generations (millions) Number of generations (millions)
Pos
terio
r pro
babi
litie
s of
20
node
s
0
20
40
60
80
100
Number of generations (millions)
Pos
terio
r pro
babi
litie
s of
20
node
s
0
20
40
60
80
100
Number of generations (millions)
Pos
terio
r pro
babi
litie
s of
20
node
s
(a) (b)
(c) (d)
0
20
40
60
80
100
0 5 10 15 20 25 30 35 40 45 50
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
182 Figure S3.3 AWTY output from the
phylogenetic analysis of Anampses.
Each bivariate plot is a pairwise
comparison of the posterior
probabilities of nodes for each of
four independent Bayesian inference
analyses as implemented in BEAST.
Low correlation among runs would
diagnose lack of convergence,
however that was not detected here.
0
20
40
60
80
100
0 20 40 60 80 100
Run
2
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
3
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
3
Run 2
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 2
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 3
183
Figure S3.4 AWTY output from the
phylogenetic analysis of Anampses.
Each plot shows the cumulative
posterior probabilities of the 15
most variable nodes for trees
searched by four independent
Markov Chain Monte Carlo
analyses as implemented in BEAST;
(a) analysis 1; (b) analysis 2; (c)
analysis 3; and (d) analysis 4. A
trend in the frequency of posterior
probabilities for nodes diagnosis
lack of convergence, but was not
observed here.
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10 0
20
40
60
80
100
0
20
40
60
80
100
0
20
40
60
80
100
Pos
terio
r pro
babi
litie
s of
15
node
sP
oste
rior p
roba
bilit
ies
of 1
5 no
des
0 1 2 3 4 5 6 7 8 9 10
0 1 2 3 4 5 6 7 8 9 10
Pos
terio
r pro
babi
litie
s of
15
node
sP
oste
rior p
roba
bilit
ies
of 1
5 no
des
0 1 2 3 4 5 6 7 8 9 10
(a) (b)
(c) (d)
Number of generations (millions)
Number of generations (millions)
Number of generations (millions)
Number of generations (millions)
184
Appendix C: Supplementary materials for Chapter 4 The following supplemental accompanies the article
Evolution of sympatric species: a case study of the coral reef fish genus Pomacanthus (Family: Pomacanthidae)
Jennifer R. Hodge 1,2,3, Charmaine I. Read 1,3, David R. Bellwood 1,2, Lynne van Herwerden 1,3
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia 3Molecular Evolution and Ecology Lab, James Cook University, Townsville, Qld, 4811, Australia
Journal of Biogeography 40(9): 1676–1687 (2013)
FIGURES
Figure S4.1 AWTY output from the phylogenetic analysis of Pomacanthus showing pairwise
comparison of the posterior probabilities of nodes for two independent Bayesian inference
analyses implemented in MRBAYES.
Figure S4.2 AWTY output from the phylogenetic analysis of Pomacanthus showing the
cumulative posterior probabilities of the 18 most variable nodes for trees searched by two
independent Markov Chain Monte Carlo analyses implemented in MRBAYES.
Figure S4.3 Inferred phylogeny of Pomacanthus obtained by Bayesian inference, maximum
likelihood and maximum parsimony analyses of three molecular loci.
185
Figure S4.4 AWTY output from the phylogenetic analysis of Pomacanthus showing pairwise
comparisons of the posterior probabilities of nodes for each of four independent Bayesian
inference analyses implemented in BEAST.
Figure S4.5 AWTY output from the phylogenetic analysis of Pomacanthus showing the
cumulative posterior probabilities of the 15 most variable nodes for trees searched by four
independent Markov Chain Monte Carlo analyses implemented in BEAST.
186
Figure S4.1 AWTY output from the
phylogenetic analysis of Pomacanthus. The
bivariate plot is a pairwise comparison of
the posterior probabilities of nodes for two
independent Bayesian inference analyses
implemented in MRBAYES. Low correlation
among runs would diagnose lack of
convergence; however, that was not
detected.
Figure S4.2 AWTY
output from the
phylogenetic analysis
of Pomacanthus. Each
plot shows the
cumulative posterior
probabilities of the 18
most variable nodes for
trees searched by two
independent Markov
Chain Monte Carlo
analyses implemented
in MRBAYES: (a)
analysis 1; and (b)
analysis 2. A trend in
the frequency of
posterior probabilities
for nodes diagnoses
lack of convergence,
but was not observed
here.
0
20
40
60
80
100
0 20 40 60 80 100
Run
2
Run 1
Pos
terio
r pro
babi
litie
s of
18
node
sP
oste
rior p
roba
bilit
ies
of 1
8 no
des
Number of generations (millions)
Number of generations (millions)
(a)
(b)
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10
187
Figure S4.3 Inferred phylogeny of Pomacanthus obtained by Bayesian inference (BI), maximum
likelihood (ML) and maximum parsimony (MP) analyses of three loci (12S, 16S and S7). The topology
shown is the best Bayesian tree with bootstrap support values (> 50%) from MP and ML (1000 and 100
bootstrap replicates, respectively) and posterior probabilities from BI (consensus of 36,002 trees).
Asterisks denote 100% support for the node across all three analyses. Sister taxa include
Chaetodontoplus mesoleucus, Chaetodontoplus duboulayi, Pygoplites diacanthus, Holacanthus tricolor,
Holacanthus limbaughi and Holacanthus africanus. Platax orbicularis was used to root the phylogeny.
P. sexstriatus
P. xanthometapon
P. arcuatus
P. navarchus
P. imperator
P. annularis
P. semicirculatus
P. chrysurus
P. asfur
P. rhomboides
P. maculosus
P. paru
H. limbaughi
H. africanus
H. tricolor
P. zonipectus
P. diacanthus
C. duboulayi
C. mesoleucus
P. orbicularis
BI/MLMP
100/9590
62
61
72
97
99
96
98
*
84/-
99/71
99/77100/97
100/93
100/97
100/97
69/-
67/-
74/-
-
-
--/-
-
*
*
*
*
188 Figure S4.4 AWTY output from the
phylogenetic analysis of
Pomacanthus. Each bivariate plot is
a pairwise comparison of the
posterior probabilities of nodes for
each of four independent Bayesian
inference analyses implemented in
BEAST. Low correlation among runs
would diagnose lack of
convergence; however, that was not
detected here.
0
20
40
60
80
100
0 20 40 60 80 100
Run
2
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
3
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 1
0
20
40
60
80
100
0 20 40 60 80 100
Run
3
Run 2
0
20
40
60
80
100
0 20 40 60 80 100
Run
4
Run 2
0
20
40
60
80
100
0 20 40 60 80 100R
un 4
Run 3
189
Figure S4.5 AWTY output from the
phylogenetic analysis of
Pomacanthus. Plots show the
cumulative posterior probabilities of
the 18 most variable nodes for trees
searched by four independent
Markov Chain Monte Carlo analyses
implemented in BEAST: (a) analysis
1; (b) analysis 2; (c) analysis 3; and
(d) analysis 4. A trend in the
frequency of posterior probabilities
for nodes diagnoses lack of
convergence, but was not observed
here.
Pos
terio
r pro
babi
litie
s o
f 18
node
s
Pos
terio
r pro
babi
litie
s o
f 18
node
sP
oste
rior p
roba
bilit
ies
of 1
8 no
des
Pos
terio
r pro
babi
litie
s o
f 18
node
s
Number of generations (millions) Number of generations (millions)
Number of generations (millions)Number of generations (millions)
(a) (b)
(c) (d)
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10 0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10
0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10 0
20
40
60
80
100
0 1 2 3 4 5 6 7 8 9 10
190
Appendix D: Supplementary materials for Chapter 5 The following supplemental accompanies the article
On the relationship between species age and geographical range in reef fishes: are widespread species older than they seem?
Jennifer R. Hodge 1,2, David R. Bellwood 1,2
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia
Global Ecology and Biogeography (2014 in press)
TABLES
Table S5.1 Median node age estimates and corresponding 95% highest posterior density
intervals from a maximum clade credibility tree constructed from post-burn-in topologies of six
independent Bayesian analyses of molecular data using nine fossil calibrations implemented in
BEAST.
FIGURES
Figure S5.1 Time-calibrated phylogeny compiled from post burn-in topologies of six
independent Bayesian analyses of molecular data using nine fossil calibrations implemented in
BEAST.
191
Figure S5.2 Plots of the log-transformed correlations for (a) all nodes in the phylogeny and
using a sister-species approach with minimum divergence time versus (b) averaged; (c)
randomised; and (d) minimum geographical range.
192
Table S5.1 Median node age estimates with corresponding 95% highest posterior density (HPD) intervals
from a maximum clade credibility tree constructed from post-burn-in topologies of six independent
Bayesian Markov Chain Monte Carlo analyses (40 × 106 generations per run) of molecular data and nine
fossil calibration points implemented in BEAST. The data set consisted of molecular sequences (16S
rRNA, 12S rRNA, CO1, cytochrome b, TMO-4C4 and S7 intron 1) for 312 Percomorph fishes used in
the multi-family phylogenetic analyses (Chapters 5–7).
Genus Species Median age (Ma) 95 % HPD (Ma) Chaetodontidae Lower bound Upper bound Amphichaetodon howensis 4.664 1.225 10.41 Amphichaetodon melbae 4.664 1.225 10.41 Chaetodon adiergastos 0.7242 0.0001546 1.985 Chaetodon argentatus 6.302 3.674 9.631 Chaetodon aureofasciatus 1.857 0.3916 3.964 Chaetodon auriga 2.209 0.8929 3.856 Chaetodon auripes 2.540 0.9352 4.240 Chaetodon austriacus 1.497 0.4979 2.896 Chaetodon baronessa 4.918 1.839 8.988 Chaetodon bennetti 6.649 4.105 9.618 Chaetodon blackburnii 7.202 4.531 10.05 Chaetodon burgessi 1.006 0.2109 2.342 Chaetodon capistratus 5.971 3.039 9.322 Chaetodon citrinellus 4.041 1.851 6.537 Chaetodon collare 3.729 2.484 5.217 Chaetodon decussatus 2.209 0.8929 3.856 Chaetodon dolosus 1.466 0.5252 2.602 Chaetodon ephippium 3.640 1.330 6.668 Chaetodon falcula 0.7941 0.2603 1.554 Chaetodon fasciatus 0.8894 0.2672 1.741 Chaetodon flavirostris 0.7242 0.0001546 1.985 Chaetodon fremblii 9.718 6.654 12.89 Chaetodon guentheri 2.013 0.9684 3.329 Chaetodon guttatissimus 1.249 0.4980 2.425 Chaetodon hoefleri 13.08 6.662 20.19 Chaetodon humeralis 9.373 6.673 12.55 Chaetodon interruptus 1.558 0.4509 3.229 Chaetodon kleinii 0.3655 0.06169 0.9483 Chaetodon larvatus 4.918 1.839 8.988 Chaetodon lineolatus 2.445 1.312 3.809 Chaetodon lunula 0.8894 0.2672 1.741 Chaetodon lunulatus 3.170 1.559 5.338 Chaetodon madagaskariensis 0.7523 0.1787 1.598 Chaetodon marleyi 10.35 6.048 15.50 Chaetodon melannotus 1.162 0.3786 2.433 Chaetodon mertensii 0.4726 0.03071 1.275 Chaetodon mesoleucos 3.946 1.410 7.141 Chaetodon meyeri 3.476 1.808 5.695 Chaetodon miliaris 1.219 0.3221 2.389 Chaetodon mitratus 1.006 0.2109 2.342 Chaetodon multicinctus 0.2885 0.06392 0.6570 Chaetodon ocellatus 10.96 7.928 14.49 Chaetodon ocellicaudus 1.162 0.3786 2.433 Chaetodon octofasciatus 1.857 0.3916 3.964 Chaetodon ornatissimus 2.732 1.325 4.617 Chaetodon oxycephalus 1.486 0.6583 2.556 Chaetodon paucifasciatus 0.4726 0.03071 1.275 Chaetodon pelewensis 0.4576 0.1276 0.9471 Chaetodon plebeius 5.250 2.913 8.133
193
Genus Species Median age (Ma) 95 % HPD (Ma) Chaetodontidae cont. Lower bound Upper bound Chaetodon punctatofasciatus 0.2885 0.06392 0.6570 Chaetodon quadrimaculatus 4.041 1.851 6.537 Chaetodon rafflesii 3.946 1.410 7.141 Chaetodon rainfordi 5.243 3.013 7.965 Chaetodon reticulatus 2.732 1.325 4.617 Chaetodon robustus 13.08 6.662 20.19 Chaetodon sanctaehelenae 1.219 0.3221 2.389 Chaetodon sedentarius 1.466 0.5252 2.602 Chaetodon selene 8.100 4.438 12.28 Chaetodon semeion 6.247 3.105 9.816 Chaetodon semilarvatus 3.276 1.820 4.940 Chaetodon speculum 3.390 1.503 5.799 Chaetodon striatus 5.971 3.039 9.322 Chaetodon tinkeri 2.135 0.3973 5.768 Chaetodon trichrous 0.3655 0.06169 0.9483 Chaetodon tricinctus 12.66 9.282 16.33 Chaetodon trifascialis 10.06 6.835 13.59 Chaetodon trifasciatus 1.497 0.4979 2.896 Chaetodon ulietensis 0.7941 0.2603 1.554 Chaetodon unimaculatus 1.558 0.4509 3.229 Chaetodon vagabundus 3.389 1.731 5.255 Chaetodon wiebeli 2.540 0.9352 4.240 Chaetodon xanthocephalus 3.640 1.330 6.668 Chaetodon xanthurus 0.7523 0.1787 1.598 Chaetodon zanzibarensis 3.390 1.503 5.799 Chelmon marginalis 1.684 0.6321 3.121 Chelmon muelleri 4.425 2.133 7.628 Chelmon rostratus 1.684 0.6321 3.121 Chelmonops curiosus 1.418 0.2735 3.619 Chelmonops truncatus 1.418 0.2735 3.619 Coradion altivelis 3.079 1.229 5.467 Coradion chrysozonus 3.079 1.229 5.467 Coradion melanopus 5.550 2.901 8.946 Forcipiger flavissimus 7.971 3.649 13.26 Forcipiger longirostris 7.971 3.649 13.26 Hemitaurichthys polylepis 0.2232 0.01580 0.6504 Hemitaurichthys thompsoni 4.429 1.967 7.732 Hemitaurichthys zoster 0.2232 0.01580 0.6504 Heniochus acuminatus 3.164 1.396 5.277 Heniochus chrysostomus 9.746 6.567 13.10 Heniochus diphreutes 3.164 1.396 5.277 Heniochus monoceros 3.274 1.503 5.633 Heniochus pleurotaenia 2.498 0.5871 5.329 Heniochus singularius 3.274 1.503 5.633 Heniochus varius 2.498 0.5871 5.329 Johnrandallia nigrirostris 14.00 9.730 18.95 Parachaetodon ocellatus 13.91 10.43 17.68 Labridae Anampses caeruleopunctatus 0.5373 0.1307 1.204 Anampses chrysocephalus 1.512 0.5776 2.841 Anampses cuvier 10.08 6.142 14.25 Anampses elegans 14.80 10.67 19.73 Anampses femininus 7.198 3.787 10.98 Anampses geographicus 19.24 13.59 26.28 Anampses lennardi 0.5373 0.1307 1.204 Anampses lineatus 4.058 2.211 6.378 Anampses melanurus 1.512 0.5776 2.841 Anampses meleagrides 7.761 4.929 11.04
194
Genus Species Median age (Ma) 95 % HPD (Ma) Labridae cont. Lower bound Upper bound Anampses neoguinaicus 7.621 4.353 11.39 Anampses twistii 7.621 4.353 11.39 Bolbometopon muricatum 17.47 12.91 22.58 Centrolabrus caeruleus 0.9127 0.2128 1.873 Centrolabrus exoletus 0.4775 0.08054 1.268 Centrolabrus trutta 0.9127 0.2128 1.873 Chlorurus atrilunula 0.6688 0.08743 1.813 Chlorurus bleekeri 2.756 1.402 4.429 Chlorurus bowersi 3.839 2.278 5.862 Chlorurus capistratoides 0.6688 0.08743 1.813 Chlorurus cyanescens 1.510 0.4230 3.209 Chlorurus enneacanthus 0.9664 0.1759 2.315 Chlorurus frontalis 0.9664 0.1759 2.315 Chlorurus gibbus 2.030 0.9095 3.641 Chlorurus japanensis 6.915 4.924 9.370 Chlorurus microrhinos 0.7725 0.09398 2.067 Chlorurus oedema 0.4659 0.04009 1.258 Chlorurus perspicillatus 9.132 5.801 13.56 Chlorurus rhakoura 0.4659 0.04009 1.258 Chlorurus sordidus 2.668 1.425 4.512 Chlorurus spilurus 2.668 1.425 4.512 Chlorurus strongylocephalus 0.7725 0.09398 2.067 Cryptotomus roseus 24.57 15.80 32.88 Diproctacanthus xanthurus 14.70 7.351 23.78 Epibulus brevis 4.286 1.035 9.545 Epibulus insidiator 4.286 1.035 9.545 Gomphosus caeruleus 3.444 1.420 7.054 Gomphosus varius 3.444 1.420 7.054 Haletta semifasciata 4.118 1.595 7.680 Hemigymnus fasciatus 7.732 3.154 14.29 Hemigymnus melapterus 7.732 3.154 14.29 Hipposcarus harid 5.944 2.535 11.62 Hipposcarus longiceps 5.944 2.535 11.62 Labrichthys unilineatus 22.36 13.80 32.22 Lachnolaimus maximus 58.63 50.00 69.86 Larabicus quadrilineatus 14.70 7.351 23.78 Leptoscarus vaigiensis 35.86 28.45 44.49 Neodax balteatus 4.118 1.595 7.680 Odax acroptilus 22.16 15.09 30.34 Odax cyanoallix 7.656 3.293 13.30 Odax cyanomelas 18.12 11.26 25.44 Odax pullus 7.656 3.293 13.30 Ophthalmolepis lineolata 31.98 24.79 41.79 Oxyjulis californica 27.74 18.66 37.36 Pseudodax moluccanus 49.39 38.73 60.77 Scarus altipinnis 2.529 1.26 4.048 Scarus arabicus 5.984 2.570 10.36 Scarus chameleon 0.7034 0.1547 1.596 Scarus coelestinus 2.682 1.203 4.621 Scarus coeruleus 3.863 1.324 6.899 Scarus collana 10.80 7.544 14.48 Scarus compressus 1.021 0.2543 2.100 Scarus dimidiatus 3.780 2.068 5.603 Scarus dubius 2.308 0.9034 4.183 Scarus falcipinnis 0.6932 0.1212 1.539 Scarus ferrugineus 1.640 0.5244 3.139 Scarus festivus 0.7034 0.1547 1.596 Scarus flavipectoralis 3.588 1.489 6.071
195
Genus Species Median age (Ma) 95 % HPD (Ma) Labridae cont. Lower bound Upper bound Scarus forsteni 4.472 2.565 6.562 Scarus frenatus 6.094 3.660 8.572 Scarus fuscopurpureus 4.245 2.259 6.637 Scarus ghobban 1.021 0.2543 2.100 Scarus globiceps 1.021 0.2543 2.100 Scarus guacamaia 0.5693 0.08921 1.409 Scarus hoefleri 4.820 2.456 7.700 Scarus hypselopterus 3.588 1.489 6.071 Scarus iseri 8.033 5.490 10.66 Scarus koputea 4.197 1.734 7.079 Scarus longipinnis 2.308 0.9034 4.183 Scarus niger 3.780 2.068 5.603 Scarus oviceps 1.380 0.4725 2.723 Scarus ovifrons 5.984 2.570 10.36 Scarus perrico 4.820 2.456 7.700 Scarus persicus 1.640 0.5244 3.139 Scarus prasiognathos 0.6932 0.1212 1.539 Scarus psittacus 9.023 6.350 12.32 Scarus quoyi 4.197 1.734 7.079 Scarus rivulatus 0.9119 0.2255 1.991 Scarus rubroviolaceus 2.669 1.132 4.502 Scarus russelii 4.245 2.259 6.637 Scarus scaber 1.380 0.4725 2.723 Scarus schlegeli 5.276 3.152 7.714 Scarus spinus 2.274 0.8596 4.215 Scarus taeniopterus 2.019 0.7149 3.791 Scarus tricolor 4.472 2.565 6.562 Scarus trispinosus 0.5693 0.08921 1.409 Scarus vetula 3.863 1.324 6.899 Scarus viridifucatus 2.274 0.8596 4.215 Scarus xanthopleura 6.467 3.824 9.364 Scarus zelindae 2.019 0.7149 3.791 Scarus zufar 7.428 4.960 10.10 Symphodus bailloni 6.715 4.505 9.375 Symphodus cinereus 7.608 5.265 10.57 Symphodus doderleini 5.989 2.426 10.02 Symphodus mediterraneus 5.096 2.914 7.792 Symphodus melanocercus 0.4775 0.08054 1.268 Symphodus melops 0.5752 0.19 1.146 Symphodus ocellatus 5.989 2.426 10.02 Symphodus roissali 0.5752 0.19 1.146 Symphodus rostratus 5.317 3.255 7.692 Symphodus tinca 5.096 2.914 7.792 Tautoga onitis 17.78 8.090 28.94 Tautogolabrus adspersus 17.78 8.090 28.94 Thalassoma amblycephalum 3.829 2.105 5.873 Thalassoma ascensionis 1.423 0.4163 3.025 Thalassoma ballieui 8.712 4.094 14.84 Thalassoma bifasciatum 3.359 1.432 5.991 Thalassoma cupido 3.375 1.720 5.356 Thalassoma duperrey 0.5995 0.1121 1.285 Thalassoma genivittatum 1.539 0.8012 2.572 Thalassoma grammaticum 0.5995 0.1121 1.285 Thalassoma hardwicke 6.871 3.871 10.19 Thalassoma hebraicum 5.328 2.895 8.484 Thalassoma jansenii 6.871 3.871 10.19 Thalassoma loxum 3.375 1.720 5.356 Thalassoma lucasanum 2.145 0.9822 3.569
196
Genus Species Median age (Ma) 95 % HPD (Ma) Labridae cont. Lower bound Upper bound Thalassoma lunare 10.12 7.312 13.37 Thalassoma lutescens 1.961 1.027 3.172 Thalassoma newtoni 3.673 1.686 6.158 Thalassoma noronhanum 3.359 1.432 5.991 Thalassoma pavo 8.459 5.079 12.42 Thalassoma purpureum 1.554 0.5377 2.961 Thalassoma quinquevittatum 9.213 5.945 12.66 Thalassoma robertsoni 2.145 0.9822 3.569 Thalassoma rueppellii 0.9532 0.4111 1.715 Thalassoma sanctaehelenae 1.423 0.4163 3.025 Thalassoma septemfasciatum 8.712 4.094 14.84 Thalassoma trilobatum 1.554 0.5377 2.961 Thalassoma virens 5.219 3.187 7.454 Xiphocheilus typus 39.55 29.81 49.99 Pomacanthidae Holacanthus africanus 8.880 5.560 13.05 Holacanthus bermudensis 0.3457 0.03093 0.9602 Holacanthus ciliaris 0.3457 0.03093 0.9602 Holacanthus clarionensis 0.1722 0.001940 0.5250 Holacanthus limbaughi 0.5038 0.1057 1.187 Holacanthus passer 0.1722 0.001940 0.5250 Holacanthus tricolor 6.649 3.905 10.16 Pomacanthus annularis 8.952 4.934 13.31 Pomacanthus arcuatus 2.398 0.9885 4.268 Pomacanthus asfur 5.759 3.645 8.297 Pomacanthus chrysurus 3.210 1.527 5.376 Pomacanthus imperator 8.952 4.934 13.31 Pomacanthus maculosus 3.948 2.210 5.884 Pomacanthus navarchus 8.757 4.776 13.51 Pomacanthus paru 2.398 0.9885 4.268 Pomacanthus rhomboides 4.870 3.007 7.040 Pomacanthus semicirculatus 3.210 1.527 5.376 Pomacanthus sexstriatus 3.274 1.340 6.018 Pomacanthus xanthometapon 3.274 1.340 6.018 Pomacanthus zonipectus 4.808 2.544 7.890 Pygoplites diacanthus 13.47 8.760 18.97 Epinephelidae Aethaloperca rogaa 27.97 20.32 37.86 Alphestes afer 4.590 2.322 7.403 Alphestes immaculatus 3.154 1.291 5.508 Alphestes multiguttatus 3.154 1.291 5.508 Anyperodon leucogrammicus 17.87 11.90 25.36 Dermatolepis dermatolepis 4.319 2.055 7.225 Dermatolepis inermis 3.112 1.268 5.564 Dermatolepis striolata 3.112 1.268 5.564 Gracila albomarginata 29.17 17.90 41.51 Mycteroperca acutirostris 0.4565 0.004133 1.553 Mycteroperca bonaci 1.618 0.6392 2.948 Mycteroperca fusca 5.787 2.247 9.765 Mycteroperca interstitialis 2.689 1.201 4.351 Mycteroperca jordani 3.580 1.899 5.512 Mycteroperca microlepis 6.888 4.758 9.403 Mycteroperca olfax 3.776 2.028 5.635 Mycteroperca phenax 0.1549 0.000001878 0.7431 Mycteroperca prionura 3.776 2.028 5.635 Mycteroperca rosacea 3.040 1.554 4.836 Mycteroperca rubra 0.4565 0.004133 1.553 Mycteroperca tigris 3.040 1.554 4.836
197
Genus Species Median age (Ma) 9 5% HPD (Ma) Epinephelidae cont. Lower bound Upper bound Mycteroperca venenosa 1.618 0.6392 2.948 Mycteroperca xenarcha 0.1549 0.000001878 0.7431 Niphon spinosus 51.32 29.49 69.90 Paranthias colonus 2.509 0.8494 5.361 Paranthias furcifer 2.509 0.8494 5.361 Plectropomus areolatus 3.691 1.579 6.102 Plectropomus laevis 3.322 1.571 5.680 Plectropomus leopardus 3.691 1.579 6.102 Plectropomus maculatus 2.957 1.485 4.681 Plectropomus oligacanthus 3.322 1.571 5.680 Plectropomus pessuliferus 2.957 1.485 4.681 Plectropomus punctatus 7.243 4.597 10.26 Saloptia powelli 23.88 13.84 36.50 Variola albimarginata 5.599 2.352 10.43 Variola louti 5.599 2.352 10.43 Fossil Calibration Taxa Acanrthurus nigricans 45.34 22.60 57.48 Ambloplites rupestris 16.58 15.80 17.96 Archoplites interruptus 16.58 15.80 17.96 Bodianus mesothorax 52.08 41.61 64.12 Calatomus carolinus 30.64 24.42 38.03 Cetoscarus bicolor 17.47 12.91 22.58 Gazza minuta 17.57 12.23 27.83 Leiognathus equulus 17.57 12.23 27.83 Luvarus imperialis 45.34 22.60 57.48 Naso lituratus 40.99 20.01 55.92 Prognathodes aculeatus 28.45 22.91 34.71 Sparisoma chrysopterum 24.57 15.80 32.88 Zanclus cornutus 40.99 20.01 55.92 Outgroup Taxa Opsanus pardus 43.56 13.93 79.73 Porichthys notatus 43.56 13.93 79.73
198
Figure S5.1 Time-calibrated phylogeny of 312 Percomoph fishes compiled from post burn-in topologies
of six independent Bayesian Markov Chain Monte Carlo analyses (40 × 106 generations per run) using
nine fossil calibration points implemented in BEAST. Molecular data included four mitochondrial loci
(16S rRNA, 12S rRNA, CO1, cytochrome b) and two nuclear loci (TMO-4C4, S7 intron 1); see Table
S2.5 in Appendix A for partitioning scheme. Nodes represent median ages from a maximum clade-
credibility tree (values provided in Table S5.1 above, along with corresponding 95% highest posterior
density intervals). Branch colours subtending nodes indicate posterior probability for the node:
black > 75%; blue 50–74%; and red < 49%. • indicates nodes calibrated with priors based on fossil data
(Table S2.4, Appendix A). Time scale is in millions of years before present.
255075100125150
Symphodus doderleini
Thalassoma hardw
icke
Scarus prasiognathos
Naso lituratus
Cha
etod
on m
adag
aska
riens
is
Chaetodon rainfordi
Chaetodon striatus
Pomacanthus annularis
Mycteroperca olfax
Pomacanthus navarchus
Cha
etod
on x
anth
urus
Chaet
odon
aus
triac
usSca
rus festi
vus
Zanclus cornutus
Thalassoma trilobatum
Anampses elegans
Cha
etod
on g
utta
tissi
mus
Thalassoma loxum
Chaetodon semilarvatus
Mycteroperca xenarcha
Thalassoma pavo
Chl
orur
us r
hako
ura
Dermatolepis striolata
Chaetodon octofasciatus
Heniochus pleurotaenia
Mycteroperca bonaci
Chl
orur
us a
trilu
nula
Pomacanthus sem
icirculatus
Cha
etod
on tr
ichr
ous
Cha
etod
on p
unct
atof
asci
atus
Symphodus bailloni
Anampses caeruleopunctatus
Chelm
on marginalis
Chaetodon ulietensis
Paranthias furcifer
Alphestes multiguttatus
Mycteroperca acutirostris
Mycteroperca prionura
Variola louti
Chaetodon melannotus
Haletta semifasciata
Chaetodon mesoleucos
Leiognathus equulus
Scar
us c
olla
na
Plectropomus areolatus
Chaetodon trici
nctus
Chaetodon oxycephalusPlectropomus pessuliferus
Scar
us ru
ssel
ii
Symphodus mediterraneus
Thalassoma am
blycephalum
Hemigym
nus melapterus
Centrolabrus caeruleus
Thalassoma robertsoni
Grammistes sexlineatus
Scar
us fu
scop
urpu
reus
Scar
us o
vifro
nsAnampses chrysocephalus
Holacanthus ciliarisSca
rus zelindae
Archoplites interruptus
Chaetodon ocellatus
Chl
orur
us s
ordi
dus
Cha
etod
on u
nim
acul
atus
Epibulus brevis
Pomacanthus chrysurus
Chaet
odon
retic
ulat
us
Mycteroperca microlepis
Chl
orur
us b
ower
si
Ambloplites rupestris
Plectropomus laevis
Diproctacanthus xanthurus
Larabicus quadrilineatus
Scar
us a
rabi
cus
Chaetodon auriga
Holacanthus limbaughi
Holacanthus passer
Thalassoma new
toni
Coradion m
elanopus
Chl
orur
us e
nnea
cant
hus
Scarus persicus
Anampses meleagrides
Saloptia powelli
Pseudodax moluccanus
Chaetodon speculum
Holacanthus africanus
Pomacanthus paru
Chaetodon trifascialis
Chaetodon decussatus
Anampses twistii
Hemigym
nus fasciatusScaru
s lo
ngip
inni
s
Chaetodon ephippium
Odax cyanomelas
Spa
risom
a ch
ryso
pter
um
Dermatolepis dermatolepis
Chaetodon rafflesii
Symphodus rostratus
Bol
bom
etop
on m
uric
atum
Scarus scaber
Hem
itaurichthys thompsoni
Variola albimarginata
Oxyjulis californica
Scarus c
hameleon
Symphodus roissali
Thalassoma ascensionis
Cha
etod
on p
elew
ensi
s
Thalassoma genivittatum
Scaru
s coe
ruleu
s
Chaetodon humeralis
Scarus r
ivulatus
Holacanthus bermudensis
Chaetodon xanthocephalus
Scarus tricolorSymphodus melanocercus
Chaetodon lineolatus
Chaetodon vagabundus
Pomacanthus im
perator
Scaru
s vetu
la
Chelm
onops truncatus
Dermatolepis inermis
Chaetodon semeion
Chaet
odon
sed
enta
rius
Cha
etod
on q
uadr
imac
ulat
us
Scarus oviceps
Chl
orur
us s
tron
gylo
ceph
alus
Anampses lineatus
Pomacanthus m
aculosus
Scarus i
seri
Scarus g
lobiceps
Opsanus pardus
Chaet
odon
orn
atiss
imus
Scar
us p
sitta
cus
Neoodax balteatus
Pomacanthus rhom
boides
Chaetodon zanzibarensis
Chl
orur
us m
icro
rhin
os
Lachnolaimus maximus
Anampses melanurus
Luvarus imperialis
Scarus viridifucatus
Holacanthus tricolor
Cha
etod
on a
rgen
tatu
s
Chl
orur
us p
ersp
icill
atus
Gom
phosus caeruleusC
hlor
urus
spi
luru
s
Chaetodon larva
tus
Pomacanthus sexstriatus
Paranthias colonus
Scaru
s per
rico
Scarus falcipinnis
Chelm
on rostratus
Thalassoma lucasanum
Chaet
odon
trifa
sciat
us
Plectropomus punctatus
Cha
etod
on m
itrat
us
Chl
orur
us b
leek
eri
Am
phichaetodon melbae
Pomacanthus xanthom
etopon
Chaetodon wiebeli
Scarus ghobban
Odax pullusOdax acroptilus
Plectropomus maculatus
Aethaloperca rogaa
Anyperodon leucogrammicus
Anampses lennardi
Chaet
odon
gue
nthe
ri
Chelm
onops curiosus
Chaetodon lunula
Scaru
s kop
utea
Pomacanthus arcuatus
Gom
phosus varius
Scarus compressus
Thalassoma sanctaehelenae
Pomacanthus asfur
Symphodus melops
Chl
orur
us o
edem
a
Scarus niger
Cha
etod
on b
urge
ssi
Chaetodon auripes
Thalassoma purpureum
Alphestes immaculatus
Thalassoma virens
Scarus altipinnis
Acanthurus nigricans
Xiphocheilus typusCratinus agassizii
Chae
todo
n m
iliar
is
Chaeto
don b
aron
essa
Holacanthus clarionensis
Forcipiger longirostris
Symphodus cinereus
Scarus taeniopterus
Leptoscarus vaigiensis
Scaru
s quo
yi
Scarus zufar
Aporops bilinearisH
eniochus singularius
Chae
todo
n sa
ncta
ehel
enae
Symphodus ocellatus
Odax cyanoallix
Chaetodon selene
Thalassoma noronhanum
Chl
orur
us fr
onta
lis
Mycteroperca jordani
Cha
etod
on m
erte
nsii
Anampses neoguinaicus
Alphestes afer
Coradion altivelis
Chaetodon flavirostris
Chaetodon falcula
Cha
etod
on fr
embl
ii
Chaetodon adiergastos
Mycteroperca rubra
Tautoga onitis
Scar
us s
chle
geli
Tautogolabrus adspersus
Scarus dimidiatus
Anampses femininus
Mycteroperca tigris
Porichthys notatusBodianus mesothorax
Epibulus insidiator
Centrolabrus trutta
Chaet
odon
mey
eri
Ophthalm
olepis lineolata
Scaru
s hoe
fleri
Chaet
odon
dol
osus
Centrolabrus exoletus
Chaet
odon
lunu
latus
Mycteroperca venenosa
Scarus spinus
Anampses geographicus
Cha
etod
on ti
nker
i
Plectropomus leopardus
Scarus coelestinusScarus xanthopleura
Mycteroperca fusca
Thalassoma septem
fasciata
Scarus trispinosus
Symphodus tinca
Mycteroperca interstitialis
Pomacanthus zonipectus
Cha
etod
on b
lack
burn
ii
Thalassoma hebraicum
Scarus forsteni
Scarus frenatus
Cha
etod
on m
ultic
inct
us
Zalanthias kelloggi
Gazza minuta
Chl
orur
us ja
pane
nsis
Chaet
odon
mar
leyi
Chaetodon aureofasciatus
Scar
us h
ypse
lopt
erus
Mycteroperca rosacea
Pygoplites diacanthus
Scarus ferrugineus
Chaetodon fasciatus
Heniochus varius
Chaetodon collare
Scar
us d
ubiu
sMycteroperca phenax
Chelm
on muelleri
Anampses cuvier
Cha
etod
on c
itrin
ellu
s
Thalassoma lunare
Thalassoma quinquevittatum
Cha
etod
on h
oefle
ri
Forcipiger flavissimus
Plectropomus oligacanthus
Chaetodon plebeius
Thalassoma ballieui
Pro
gnat
hode
s ac
ulea
tus
Thalassoma cupido
Chlorurus gibbus
Scar
us fl
avip
ecto
ralis
Labrichthys unilineatus
Chaetodon capistratus
Parachaetodon oce
llatus
Niphon spinosus
Coradion chrysozonus
Gracila albomarginata
Chaetodon ocellicaudus
Scarus rubroviolaceus
Chl
orur
us c
apis
trat
oide
s
Scarus guacamaia
Chaetodon bennetti
Cha
etod
on p
auci
fasc
iatu
s
Cha
etod
on in
terr
uptu
s
Cha
etod
on k
lein
ii
Cha
etod
on r
obus
tus
Hen
ioch
us d
iphr
eute
sH
enio
chus
acu
min
atus
Hen
ioch
us m
onoc
eros
Heniochus chrysostom
us
Johnrandallia nigrirostris
Hem
itaurichthys polylepis
Hem
itaurichthys zoster
Am
phichaetodon howensis
Chl
orur
us c
yane
scen
sC
etos
caru
s bi
colo
rH
ippo
scar
us lo
ngic
eps
Hip
posc
arus
har
idC
rypt
otom
us r
oseu
sC
alot
omus
car
olin
us
Thalassoma duperrey
Thalassoma gram
maticum
Thalassoma rueppellii
Thalassoma lutescens
Thalassoma jansenii
Thalassoma bifasciatum
ChaetodontidaePomacanthidae
Serra
nida
e
Labridae
199
Figure S5.2 All analytical approaches were repeated using log-transformed data and linear regression.
Linear models were not plotted unless the relationship was significant. (a) No significant linear log-log
relationship was detected between geographical range and species age across all nodes in the phylogeny
(r 2 = 0.00544; n = 293; d.f. = 1, 291; F = 1.59; P = 0.208). (b) No significant linear log-log relationship
was detected between average geographical range area (calculated as the mean of two sister-species
ranges) and species’ age for all sister-pairs (r 2 = 0.0247; n = 91; d.f. = 1, 89; F = 2.26; P = 0.137). (c)
Using a sister-species approach and randomly selecting one of two geographical ranges per sister-pair,
this plot shows the relationship between the log-transformed variables for the most significant
randomised replicate (9.3% of 1000 bootstrapped replicates were significant; the range of statistical
values across all 93 significant replicates: r 2 = 0.0993–0.0427; n = 91; d.f. = 1, 89; F = 9.82–3.97; P =
0.00234–0.0495). (d) A marginally non-significant linear log-log relationship was detected between
minimum geographical range and species age (r 2 = 0.0419; n = 91; d.f. = 1, 89; F = 3.89; P = 0.0516). If a
type II error produced a false non-significant result in this analysis and there were a significant
relationship between the log-transformed data, it would suggest that as species age their geographical
ranges increase exponentially. Such a pattern may be consistent with the ‘stasis post-expansion’ or ‘taxon
cycle’ theoretical models of geographical range evolution if range expansion occurs rapidly.
1112
1314
15
Log-divergence time–2 –1 0 1 2 3 4
16Lo
g-ge
ogra
phic
al ra
nge
(a) (b)
17
(c)
Min
imum
log-
geog
raph
ical
rang
e11
1213
1415
16
–2 –1 0 1 2Log-minimum divergence time
(d)
Log-
geog
raph
ical
rang
e4.
55.
05.
56.
06.
57.
0
–0.5 0.0 0.5 1.0Log-minimum divergence time
Ave
rage
log-
geog
raph
ical
rang
e
Log-minimum divergence time
4.5
5.0
5.5
6.0
6.5
7.0
–0.5 0.0 0.5 1.0
200
Appendix E: Supplementary materials for Chapter 6 The following supplemental accompanies the article
Temporal evolution of coral reef fishes: global patterns and disparity in isolated locations
Jennifer R. Hodge 1,2,3, Lynne van Herwerden 1,3, David R. Bellwood 1,2
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia 3Centre for Sustainable Tropical Fisheries and Aquaculture, James Cook University, Townsville,
Qld, 4811, Australia
Journal of Biogeography 41(11): 2115–2127 (2014)
TABLES
Table S6.1 The number of species per genus and the proportion of missing extant species in the
multi-family phylogenetic tree.
REFERENCE LIST
201
Table S6.1 The total number of extant, nominal species for each genus included in the multi-family
phylogenetic analyses; the number of extant, nominal species missing from the multi-family phylogenetic
analyses, and the resultant proportion of missing extant, nominal species for each genus. Species
descriptions were obtained from the IUCN Red List (IUCN, 2011) and FishBase
(http://www.fishbase.org/).
Genus Total No. constituent species
No. missing extant species
Percent missing extant species
Chaetodontidae Amphichaetodon 2 0 0 Chaetodon 87 15 17.2 Chelmon 3 0 0 Chelmonops 2 0 0 Coradion 3 0 0 Forcipiger 2 0 0 Hemitaurichthys 4 1 25.0 Heniochus 8 1 12.5 Jonhrandallia 1 0 0 Parachaetodon 1 0 0 Labridae Anampses 12 0 0 Bolbometopon 1 0 0 Centrolabrus 3 0 0 Chlorurus 18 2 11.1 Cryptotomus 1 0 0 Diproctacanthus 1 0 0 Epibulus 2 0 0 Gomphosus 2 0 0 Haletta 1 0 0 Hemigymnus 2 0 0 Hipposcarus 2 0 0 Labrichthys 1 0 0 Lachnolaimus 1 0 0 Larabicus 1 0 0 Leptoscarus 1 0 0 Neodax 1 0 0 Odax 4 0 0 Ophthalmolepis 1 0 0 Oxyjulis 1 0 0 Pseudodax 1 0 0 Scarus 52 6 11.5 Symphodus 10 0 0 Tautoga 1 0 0 Tautogolabrus 1 0 0 Thalassoma 28 2 7.1 Xiphocheilus 1 0 0 Pomacanthidae Holacanthus 8 1 12.5 Pomacanthus 13 0 0 Pygoplites 1 0 0 Epinephelidae Aethaloperca 1 0 0 Alphestes 3 0 0 Anyperodon 1 0 0 Aporops 1 0 0 Dermatolepis 3 0 0 Cratinus 1 0 0 Gracila 1 0 0
202
Genus Total number of constituent species
Number of missing extant species
Percentage of missing extant species
Epinephelidae cont.
Grammistes 1 0 0 Mycteroperca 15 1 6.7 Niphon 1 0 0 Paranthias 2 0 0 Plectropomus 7 1 14.3 Saloptia 1 0 0 Variola 2 0 0 Zalanthias 1 0 0
References
IUCN (2011) IUCN Red List of Threatened Species. Version 2011.2. Available at:
http://www.iucnredlist.org/.
203
Appendix F: Supplementary materials for Chapter 7 The following supplemental accompanies the article
The geography of speciation in coral reef fishes
Jennifer R. Hodge 1,2, David R. Bellwood 1,2
1School of Marine and Tropical Biology, James Cook University, Townsville, Qld, 4811, Australia 2Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University,
Townsville, Qld, 4811, Australia
Submitted for publication
FIGURES
Figure S7.1 Map showing allopatric spits that did not show congruence and/or align with a
single previously described biogeographical barrier.
REFERENCE LIST
204
Figure S7.1 Allopatric spits classified as ‘other’ (n = 13) that did not show congruence and/or align with a single previously described biogeographical barrier are mapped as
grey lines. The area of likely vicariance was ambiguous for most of these splits because the distance between their sister-species distributions was large enough to span
multiple biogeographical barriers (see Figure 7.2a for location of all previously described barriers considered herein). However, some of the remaining splits did solely
correspond to previously described barriers, including: II, n, o, and q (cf. Bellwood & Wainwright, 2002; Rocha et al., 2007; Floeter et al., 2008; Luiz et al., 2012). The
average degree of symmetry for these splits is plotted in Figure 7.2c, and was not significantly different from a random distribution of mean values drawn from the full pool
of allopatric sister-species.
o
II
n q
205
References
Bellwood, D.R. & Wainwright, P.C. (2002) The history and biogeography of fishes on coral
reefs. Coral Reef Fishes: Dynamics and Diversity in a Complex System (ed. by P.F.
Sale), pp. 5–32. Academic Press, New York.
Luiz, O.J., Madin, J.S., Robertson, D.R., Rocha, L.A., Wirtz, P. & Floeter, S.R. (2012)
Ecological traits influencing range expansion across large oceanic dispersal barriers:
insights from tropical Atlantic reef fishes. Proceedings of the Royal Society B:
Biological Sciences, 279, 1033–1040.
Rocha, L.A., Craig, M.T. & Bowen, B.W. (2007) Phylogeography and the conservation of coral
reef fishes. Coral Reefs, 26, 513–513.
206
Appendix G: Publications arising from thesis
Hodge, J. R., Read, C. I., van Herwerden, L. & Bellwood, D. R. (2012) The role of peripheral
endemism in species diversification: evidence from the coral reef fish genus Anampses
(Family: Labridae). Molecular Phylogenetics and Evolution 62(2), 653–663. doi:
10.1016/j.ympev.2011.11.007
Hodge, J. R., Read, C. I., Bellwood, D. R. & van Herwerden, L. (2013) Evolution of sympatric
species: a case study of the coral reef fish genus Pomacanthus (Family:
Pomacanthidae). Journal of Biogeography 40(9), 1676–1687. doi: 10.1111/jbi.12124
Hodge, J. R., van Herwerden, L. & Bellwood, D. R. (2014) Temporal evolution of coral reef
fishes: global patterns and disparity in isolated locations. Journal of Biogeography
41(11), 2115–2127. doi: 10.1111/jbi.12356
Hodge, J. R., & Bellwood, D. R. (2014 in press) On the relationship between species age and
geographical range in reef fishes: are widespread species older than they seem? Global
Ecology and Biogeography.
Hodge, J. R., & Bellwood, D. R. (2014 submitted) The geography of speciation in coral reef
fishes.
The role of peripheral endemism in species diversification: Evidencefrom the coral reef fish genus Anampses (Family: Labridae)
Jennifer R. Hodge a,b,c,⇑, Charmaine I. Read a,c, Lynne van Herwerden a,c, David R. Bellwood a,b
a School of Marine and Tropical Biology, James Cook University, Townsville, Qld 4811, AustraliabAustralian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Qld 4811, AustraliacMolecular Evolution and Ecology Lab, James Cook University, Townsville, Qld 4811, Australia
a r t i c l e i n f o
Article history:Received 24 July 2011Revised 9 November 2011Accepted 9 November 2011Available online 23 November 2011
Keywords:Evolutionary historyMolecular phylogenyAge estimationSuccessive divisionSuccessive colonisationPeripheral budding
a b s t r a c t
We examined how peripherally isolated endemic species may have contributed to the biodiversity of theIndo-Australian Archipelago biodiversity hotspot by reconstructing the evolutionary history of thewrasse genus Anampses. We identified three alternate models of diversification: the vicariance-based‘successive division’ model, and the dispersal-based ‘successive colonisation’ and ‘peripheral budding’models. The genus was well suited for this study given its relatively high proportion (42%) of endemicspecies, its reasonably low diversity (12 species), which permitted complete taxon sampling, and itswidespread tropical Indo-Pacific distribution. Monophyly of the genus was strongly supported by threephylogenetic analyses: maximum parsimony, maximum likelihood, and Bayesian inference based onmitochondrial CO1 and 12S rRNA and nuclear S7 sequences. Estimates of species divergence times fromfossil-calibrated Bayesian inference suggest that Anampses arose in the mid-Eocene and subsequentlydiversified throughout the Miocene. Evolutionary relationships within the genus, combined with limitedspatial and temporal concordance among endemics, offer support for all three alternate models of diver-sification. Our findings emphasise the importance of peripherally isolated locations in creating and main-taining endemic species and their contribution to the biodiversity of the Indo-Australian Archipelago.
� 2011 Elsevier Inc. All rights reserved.
1055-7903/$ - see front matter � 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.ympev.2011.11.007
Molecular Phylogenetics and Evolution 62 (2012) 653–663
Contents lists available at SciVerse ScienceDirect
Molecular Phylogenetics and Evolution
journal homepage: www.elsevier .com/ locate /ympev
ORIGINALARTICLE
Evolution of sympatric species:a case study of the coral reef fishgenus Pomacanthus (Pomacanthidae)Jennifer R. Hodge1,2,3*, Charmaine I. Read1,3, David R. Bellwood1,2
and Lynne van Herwerden1,3
1School of Marine and Tropical Biology,
James Cook University, Townsville, QLD,
4811, Australia, 2Australian Research Council
Centre of Excellence for Coral Reef Studies,
James Cook University, Townsville, QLD,
4811, Australia, 3Molecular Evolution and
Ecology Lab, James Cook University,
Townsville, QLD, 4811, Australia
*Correspondence: Jennifer R. Hodge, School of
Marine and Tropical Biology, James Cook
University, Townsville, QLD 4811, Australia.
E-mail: [email protected]
ABSTRACT
Aim To establish a spatial and temporal framework within which we can begin
to investigate the role of geography in the evolution of Pomacanthus species.
This study examines the phylogenetic relationships among Pomacanthus spe-
cies, tests whether the degree of sympatry among sister taxa correlates with
their age, and explores potential modes of diversification.
Location Pan-tropical coral reef systems.
Methods Three gene regions (12S rRNA, 16S rRNA and nuclear S7) from all
13 Pomacanthus species were used in conjunction with fossil calibration to
reconstruct a chronogram. IUCN maps were used to evaluate geographical
range overlap. Various age–range correlation (ARC) analyses were used to test
for a correlation between range overlap and node age. Range-size symmetry
and node age were also examined.
Results Biogeographical splits within the phylogeny corroborated key biogeo-
graphical events well, suggesting a potential role for allopatric speciation in the
evolutionary history of the genus. ARC analyses suggested that speciation lead-
ing to sympatry is widespread in Pomacanthus, with 80% of sister species
showing complete or substantial (> 85%) range overlap. No significant rela-
tionship between degree of sympatry and node age was recovered, which dem-
onstrates that, for this group, exceptional sympatry is not necessarily a result
of relatively old lineages. Range-size symmetry analysis was consistent with
models of peripatric speciation within a finite area.
Main conclusions We present evidence consistent with allopatric speciation,
while our ARC results are consistent with simulations of sympatric speciation,
with the possibility of peripheral budding having led to the richness of sympat-
ric species in the West Indian Ocean. We discuss how these processes can
affect the interpretation of temporal and spatial analyses, and provide a basis
for future investigations into the processes driving speciation.
Keywords
Age–range correlation, allopatry, coral reefs, divergence dating, fossil calibra-
tion, marine biogeography, pan-tropical, peripheral budding, speciation mode,
sympatry.
1676 http://wileyonlinelibrary.com/journal/jbi ª 2013 John Wiley & Sons Ltddoi:10.1111/jbi.12124
Journal of Biogeography (J. Biogeogr.) (2013) 40, 1676–1687
ORIGINALARTICLE
Temporal evolution of coral reef fishes:global patterns and disparity in isolatedlocationsJennifer R. Hodge1,2,3*, Lynne van Herwerden1,3 and David R. Bellwood1,2
1School of Marine and Tropical Biology, James
Cook University, Townsville, QLD 4811,
Australia, 2Australian Research Council
Centre of Excellence for Coral Reef Studies,
James Cook University, Townsville, QLD 4811,
Australia, 3Centre for Sustainable Tropical
Fisheries and Aquaculture, James Cook
University, Townsville, QLD 4811, Australia
*Correspondence: Jennifer R. Hodge, School of
Marine and Tropical Biology, James Cook
University, Townsville, QLD 4811, Australia.
E-mail: [email protected]
ABSTRACT
Aim Many coral reef regions have a history of isolation and extinction. Our
aim was to test whether the disparate evolutionary and biogeographical histo-
ries of the world’s coral reef regions have significantly impacted temporal pat-
terns of speciation within regions. In essence, do assemblages in peripheral
locations contain the youngest coral reef fish species?
Location Pan-tropical coral reef systems.
Methods Molecular data (mitochondrial 16S rRNA, 12S rRNA, CO1, cyto-
chrome b; nuclear TMO-4C4, S7 intron 1) were assembled for genera with
near-complete taxon sampling (minimum 70% nominal species) from four
major coral reef fish families (Chaetodontidae, Labridae, Pomacanthidae and
Serranidae). This was combined with fossil data to simultaneously infer the
phylogeny and estimate species’ divergence times. Species’ distributions were
quantified using IUCN maps and the ages of species with different biogeo-
graphical extents were compared. Model fitting was used to compare the distri-
bution of species’ ages across the whole phylogeny with age distributions of
species restricted to the Red Sea and Hawaiian Islands.
Results Temporal patterns of coral reef fish divergence were similar among
major marine realms and regions. However, notable differences were recorded
between the Red Sea and Hawaiian Islands. Red Sea endemics have diverged
consistently throughout the last 16 Myr, whereas endemic species colonized the
Hawaiian Islands in two distinct waves (0–3 Ma and 8–12 Ma). Differences in
the proportions of allopatric and sympatric sister-species between Red Sea and
Hawaiian endemics were also detected.
Main conclusions Despite differing geological histories, marine realms and
regions have all experienced comparable and relatively recent divergences of
extant coral reef fish species. Differences in age distributions and spatial rela-
tionships of endemic species in the Red Sea and Hawaiian Islands suggest that
markedly different processes have shaped patterns of diversification in these
peripherally isolated locations.
Keywords
Biodiversity, coral reef fishes, Coral Triangle, endemism, evolutionary age,
Hawaiian Islands, marine biogeography, peripatric speciation, Red Sea, sister-
species.
ª 2014 John Wiley & Sons Ltd http://wileyonlinelibrary.com/journal/jbi 2115doi:10.1111/jbi.12356
Journal of Biogeography (J. Biogeogr.) (2014) 41, 2115–2127